KR20040066517A - Device and method for outputting a private image using a public display - Google Patents

Abstract

PURPOSE: An apparatus and a method for outputting a private image using a display open to the public are provided to generate effective masking image considering visual perception characteristic of a human being. CONSTITUTION: A private image generating element generates a private image. A masking image generating element(326) generates a masking image masking the private image. An image data frame sequence generating element(322) generates a sequence pattern for the private image and the masking image. A video controller(312) outputs the private image and the masking image to a display(304) according to the sequence pattern. The masking image generating element generates a dynamic inverse image of the private image as a masking image according to refresh speed of a splay and the image sequence pattern.

Description

DEVICE AND METHOD FOR OUTPUTTING A PRIVATE IMAGE USING A PUBLIC DISPLAY}

The present invention relates to an apparatus and a method for outputting a private image, and more particularly, to an apparatus and a method for outputting a private image so that a non-licensed person cannot see in a public display.

Portable terminal devices such as mobile phones, personal digital assistants (PDAs), notebook computers, and desktop personal computers (PCs) are frequently used in public places. At this time, the contents of the display monitor can be seen by everyone within the viewing distance of the display. Because of this security issue, when you use your computer to write text, mail, chat, watch videos, and so on, there are many restrictions on using your computer. In addition to the use of personal computers, privacy issues can also arise when companies, governments, and others work on computers with confidential documents. In addition, display security issues exist in various fields. For example, an automatic teller machine (hereinafter referred to as "ATM") is placed in a public place, so secret information such as password keystrokes and transaction details of ATM users can be easily exposed. Therefore, it is useful to develop a display that provides private information to authorized users on a publicly viewable monitor, while an unauthorized person cannot view private information on the same monitor.

Early liquid crystal display (hereinafter referred to as "LCD") monitors have a narrow viewing angle, which causes the screen to appear dark when viewed slightly sideways from the front. Since this point of view was inconvenient in terms of general use, technological advances have been made in the direction of widening the viewing angle, but from a security point of view, it has some security and is the most basic form of private display. Can be. A display with a micro blocker that further develops this concept is disclosed in US Pat. No. 5,528,319 (the inventor is Russel). However, this method has a disadvantage in that the contents of the display are completely exposed to the person behind the user.

A more advanced concept of private display is the security LCD released by Scepter in 1998. This product using polarized light removes the polarizing plate inside an ordinary LCD monitor and allows the user to use polarized glasses. However, this product is inconvenient for general use because privacy can be perceived only when privacy is not important. In addition, unlicensed people who wear glasses with simple polarization characteristics, such as general sunglasses, can see the contents, so the security is low. Eventually the product failed and was discontinued in the market. With a slight refinement of this approach, in 2001, MMI (http://www.man-machine.com/invisivw.htm) introduced a product that can be attached to or detached from a polarizer. However, this product also has the disadvantage that unauthorized users wearing simple polarized glasses such as sunglasses can view the contents.

The most complete private display is a head-mounted display (hereinafter referred to as "HMD"). However, HMD has a problem that the display and the optical system are both contained in glasses, which is expensive, heavy to wear, and high in power consumption.

Although not a private display, it is a technically relevant display, and a multi-screen display in which one display displays two different images and a user with shutter glasses visually recognizes their own image is disclosed in Korean Patent Application No. 1991-0000391 ( Applicant is Samsung Electronics, the name of the invention "Multi-screen display and visual perception device of the monitor" and Korea Patent Application 1997-044686 (Applicant is Samsung Electronics, the name of the invention "Two screen simultaneous visual imaging device" And Korean Patent Application No. 1999-0051191 (Applicant Choi Hyo-seung, the name of the invention is "Multi-Video simultaneous playback device"). In addition, a technique of applying multi-screen display to a game is disclosed in US Patent No. 5,963,371, assigned to Intel. Multi-screen display is not a private display because people other than those who wear shutter glasses and see the display can see the contents of the display to some extent. In other words, the shutter is only worn to block other images, and there is no means to protect the private images.

In the present specification, one display screen distinguished by vertical sync of a monitor is called a monitor frame, and one piece of image data is called an image data frame. One image data frame may be the same size or different from one frame of the monitor. A private image (hereinafter referred to as "P image") is a private private image of an authorized user. The masking image (hereinafter, referred to as "M image") is an image that blocks unauthorized users from viewing a private image of an authorized user.

Private displays are now being implemented that use shutter glasses to protect private information. It is considered to be the most competitive way at present because of its low cost, light weight, and plenty of room for performance. Private displays must meet all three performance criteria: user visual perception, bare-eye security, and anti-spy security. 'User perception performance' means that the authorized user can see the image clearly without any visual discomfort or fatigue, and 'eye-eye security performance' prevents the unauthorized person without the shutter from seeing the image clearly. Security against counterspy 'is to prevent unauthorized or intentional spypers with shutters from seeing the images clearly.

In the "synchronous" method of Sun Microsystems, disclosed in US Patent No. 5,629,984, a private image data frame (hereinafter referred to as "P frame") and a masking image data frame (masking image data frame) (Hereinafter referred to as " M frame ") is displayed while being replaced according to the frame synchronization of the monitor, and at the same time, the shutter glasses are opened and closed so that only the user who owns the shutter glasses can see the private image. The replacement of the P frame and the M frame is performed at a ratio of 1: m (m = 1, 2, 3, ...) which displays the M frame every time the P frame is displayed one time. This method satisfies 'eye security performance', but does not consider 'anti-spy security performance', and as m increases, 'user visual perception performance' also decreases. It is not effective to cover the private image by generating a simple white flash image with the M frame masking image. Whenever the shutter needs to be opened and closed, it is easy for the spy to intercept the shutter by sending the shutter open / closed signal without encryption.

In IBM's "Asynchronous" method disclosed in GB 2360414A, only P / M data frames and shutter glasses are synchronized while displaying P frames and M frames without matching the frame synchronization of the monitor. In this method, the display time of the P / M image data frame is variable and encrypted, thereby improving 'security against spoilers'. In addition, by encrypting and transmitting the shutter open / close signal to the shutter under the premise of the timer on the shutter side, the security performance against the spy explorer was enhanced to prevent the spy from intercepting the shutter open / close signal. However, due to the asynchronous of the image frame and the monitor frame, 'user visual performance' is reduced. It is inconvenient for the user to see because the light intensity density varies according to the area of the monitor, and it is inconvenient because image unevenness occurs at the boundary where the P and M images are switched due to the combination of the asynchronous characteristics and the finite response time of the shutter. In addition, the P-display is more likely to be displayed in a specific area (top) of the monitor, thereby degrading 'eye-eye security' and 'anti-spy security'. In addition, since the image data frame repeatedly replaces P and M, the probability that the prober can view the P image through tuning is increased. In addition, since masking image data of M frames is generated as a simple random pattern image, it is not effective to cover a private image. There is also room for improvement in that a timer is required on the receiving side for the encrypted transmission of the shutter open / close signal.

The white flash masking image disclosed in US Pat. No. 5,629,984 or the random masking image disclosed in GB 2360414 is difficult to effectively block private images. Since the uniform white flash masking image has a low spatial frequency and a low temporal frequency, it is difficult to mask a private image having a different temporal frequency, spatial frequency, or perceptual grouping. In order for masking to be effective with flash masking images, multiple flash masking images must be displayed for one private image, so that the user's visual perception performance of the private image is poor. The random masking image has a wider temporal frequency band and a spatial frequency band for luminance and color, respectively, so that masking performance is superior to that of the flash masking image, but since a human does not distinguish between a sensitive time frequency and a spatial frequency band and an insensitive frequency band. Due to this, it is difficult to mask a general private video with a lot of signals of a specific frequency band. In addition, it is difficult to mask perceptually grouped private images with cognitive meaning as random masking images without cognitive meaning.

The private display of the Mistubishi Electric Research Laboratory (MERL), which was launched on February 2002 at the website http://www.merl.com/papers/TR2002-11/ , is synchronous. In the private display of MERL, the inverse image of the private image data is generated and displayed as masking image data by using the time integration of the eyes, thereby allowing the unauthorized user to see the uniform gray image, which is the average image of the private image and the inverse image. Security performance ”. Randomly generate and provide P / M video frame sequence to improve 'anti-spy security'. In particular, the use of expensive fast shutters such as FLC suggests pixel-based shuttering rather than frame-based shuttering. In the MERL method, the gamma of the monitor was taken into account when generating a reversed phase, but an incomplete reversed phase was generated due to a lack of understanding of human visual perception. In addition, it takes a lot of time to calculate the inverse phase every frame and thus cannot be a real-time system.

In the MERL method, the gamma of the monitor was taken into account when generating the reverse phase, but the incomplete phase was generated due to the lack of consideration of the human visual perception. And because it takes a lot of time to calculate the reverse phase every frame, it could not be a real-time system, so it could be applied only to the private display of still images. In the private display of MERL, disturbed images with cognitive meaning were used as masking images. In this case, the dynamic range of the closed image is reduced compared to the masking image to make the disturbing image more visible. The dynamic range is a concept such as the difference between the maximum value and the minimum value of the color space, the range of monitor brightness, and the range of voltage to the monitor. However, since the MERL method used a specific photographic image as a disturbing image, which is effective only to cover a specific still closed image, it is not a systematic and strategic disturbance image generation method that can be used for a general closed image and does not consider human visual perception characteristics. In addition, it could not be applied to provide a disturbing image in real time every frame, so it could be applied only to the private display of still images.

The present invention has been proposed to solve the problems of the prior art, and an object of the present invention is to generate a masking image that is more effective for masking a closed image in consideration of human visual perception characteristics.

Another object of the present invention is to generate a masking image covering a closed image in real time.

1 is a system configuration of a notebook embodiment to which the present invention is applied.

2 is a block diagram of a shutter opening and closing means according to the present invention.

3 is a system configuration diagram of a software-only driver embodiment according to the present invention.

4 is a block diagram of an example of the video controller shown in FIG.

5 is a system diagram of a software based application level embodiment of the present invention.

6 is a configuration diagram of an example of a wired and wireless communication interface between a computer body and shutter opening and closing means.

7 is a system configuration diagram of a hardware type external controller embodiment according to the present invention;

8 is a block diagram of an example of the external controller shown in FIG.

9 is a system diagram of a completely independent hardware type external controller embodiment according to the present invention;

10 is a system configuration diagram of a software and hardware integrated integrated controller embodiment according to the present invention.

11 is a block diagram of a circuit for generating a masking image using a nonlinear inverter in accordance with the present invention.

FIG. 12A is a diagram showing a spatial versus sensitivity function for luminance and color of human visual perception, and FIG. 12B is a diagram showing a time versus sensitivity function for luminance and color of human visual perception.

13 illustrates the time delay between the central and peripheral mechanisms of the human optic nerve.

14 illustrates a general process of image generation and perception.

15 illustrates an exemplary monitor transfer function.

FIG. 16 illustrates a typical human visual perceptual transfer function. FIG.

17 illustrates a typical monitor pixel response.

18 shows an example of a pattern used to generate a dynamic reversed phase by the dynamic pattern experiment method of the present invention.

Fig. 19 is a diagram for explaining a state of a general color table.

20 is a view for explaining a color table changing method according to the present invention.

21 is a flowchart illustrating a method of generating and managing a disturbing image according to the present invention.

Fig. 22 shows the spatial and temporal frequency characteristics of human visual neurons, human visual angles, and typical image data.

23 illustrates a CIE Lab color space.

24 illustrates a human visual perception process.

25 is a diagram illustrating a process of generating a disturbing image by combining image elements according to the present invention.

FIG. 26 is a diagram for explaining a P / M image sequence and a shutter opening / closing sequence generation process according to the present invention; FIG.

According to an aspect of the present invention, there is provided an apparatus for outputting a private image using a public display, comprising: means for generating a private image, means for generating a masking image for masking the private image, and Means for generating an image sequence pattern for the private image and the masking image, and means for outputting the private image and the masking image to the display according to the image sequence pattern, wherein the masking image generating means includes: According to the refresh rate and the image sequence pattern, the dynamic reverse phase of the closed image is generated as a masking image.

In another aspect, the present invention provides an apparatus for outputting a private image using a public display, comprising: means for generating a private image, means for generating a masking image for masking the private image, and for the private image and the masked image Means for generating an image sequence pattern, and means for outputting the private image and the masking image to the display according to the image sequence pattern, wherein the masking image generating means masks a disturbing image based on human visual perception characteristics. It is another feature to generate the image.

In another aspect, the present invention provides a device for outputting a private image using a public display, comprising: means for receiving a private image from an external server or generating a private image by itself, means for receiving a masking image from an external server, And means for generating an image sequence pattern for the image and the masked image, and means for outputting the private image and the masked image to the display according to the image sequence pattern.

In addition, the present invention provides a device for outputting a private image using a public display, means for receiving a private image from the outside, means for generating a masking image for masking the private image, the private image and the masked image And means for generating an image sequence pattern for and a means for outputting the private image and the masking image to the display according to the image sequence pattern.

In addition, the present invention provides a device for outputting a private image using a public display, means for receiving a private image from the outside, means for generating a masking image for masking the private image, the private image and the masked image And means for receiving an image sequence pattern for and a means for outputting the private image and the masking image to the display according to the image sequence pattern.

In another aspect, the present invention provides an apparatus for outputting a private image using a public display, comprising: means for generating a private image, means for generating a masking image for masking the private image, and for the private image and the masked image Means for generating an image sequence pattern, and means for outputting the private image and the masking image to the display according to the image sequence pattern, wherein the masking image generating means comprises a color table having a color table for generating a masking image; The apparatus may further include a storage unit and an image converter configured to generate a masked image by referring to the masking image generation color table.

Hereinafter, with reference to the accompanying drawings will be described embodiments of the present invention; In the drawings, the same reference numerals are used to indicate the same or similar elements or signals in the drawings.

[System structure]

Private displays can be widely used in various monitor display devices such as desktop PCs, notebooks, PDAs, mobile phones, TVs, DVDs, ATM / CD machines, and door lock information input devices. A description will be given of an embodiment of a PC monitor, which is the most typical of these. Those skilled in the art can easily change and apply to other display devices. In general, other display devices have a simpler structure than a PC monitor, and thus may implement a private display without some of the essential components of the PC monitor embodiment.

1 is a basic configuration diagram of an embodiment applied to a notebook PC. As shown, a computer device 104 having a monitor 102, shutter opening and closing means 106 for filtering light, wired and wireless communication means 108 for connecting the computer 104 and the shutter opening and closing means 106. , Private display software (not shown). Computer 104 may include private display software stored in computer readable memory. The computer 104 displays a private image and a masking image for masking the private image on the monitor 102 by the user's request or by itself, and transmits a corresponding shutter opening / closing signal to the shutter opening / closing means 106 to open and close the shutter. Activating the means 106 ensures that only authorized users can view private images. The computer 104 refers to various information devices that display and operate an image on a monitor such as a desktop PC, a notebook, a PDA, a mobile phone, a TV, a DVD, an ATM / CD device, a door lock information input device, and the like.

The shutter opening and closing means 106 may be mechanical or photoelectric such as a liquid crystal shutter. The shutter opening and closing means 106 may be manufactured in various forms such as shutter glasses having one or several shutter lenses, a shutter structure having a support, and a shutter cap. 1 shows shutter opening and closing means in the form of shutter glasses.

When the computer 104 is used in a bright environment, the ambient light reflected from the monitor 104 may be incident on the user's eye, resulting in a low contrast of the monitor image. In one embodiment, an ambient light filter (not shown), such as “3M ^™ Privacy ComputerFilters” sold by 3M, is attached to the front of the monitor 102. The ambient light blocking filter is a light filter whose light transmittance depends on the angle of incidence. The ambient light filter consists of a film with a microlouver blocker that functions as a sunscreen window blind, blocking out incident light from above a certain angle. In another embodiment, a brightness enhancement film, such as "Vikuiti ^™ Display Enhancement Films" sold by 3M, is attached to the front of the monitor 102 to improve the brightness of the monitor. When the filter is attached to the front surface of the monitor 102, it is effective to block the ambient light incident and reflected on the monitor 102, but cannot directly block the ambient light incident on the shutter opening and closing means 106. In another embodiment of the present invention, the ambient light blocking filter 202 is attached to the shutter opening and closing means 106 in order to block the ambient light directly incident to the shutter opening and closing means 106. When the ambient light blocking filter 202 is attached, the contrast of the image perceived by the user increases, but the overall brightness decreases slightly. In another embodiment, it is possible to mount the filter 202 in bright places and to detach it in dark places.

3 is a configuration diagram of a system using a software-based dedicated driver according to an embodiment of the present invention. The software method means that the software implements functions except for the shutter opening and closing means 306 and the transmission / reception interface unit 308. Here, the dedicated driver 310 refers to a driver that accesses the video controller 312 such as a graphics card separately from the graphics driver 314 in the PC 302 to implement a private display in real time.

The private display control block 318 includes a security performance control unit, an encryption unit, a user authentication unit, and a management unit, and authenticates a user from the user interface 320, and displays a display security level according to an authorized user's authentication level and a user's input. Set up and manage. The user authentication method may be authenticated by receiving an identification number (hereinafter referred to as “ID”) and a password input of the user from the user interface 320. In addition, user authentication can be performed by connecting an authorized shutter opening and closing means 306 without inputting an ID and a password. In addition, the user can be authenticated by connecting the permitted shutter opening and closing means 306 and receiving input of the authorized user ID and password. Authorization and activation of the shutter opening and closing means is performed with a serial number of a product embedded in a read only memory (ROM) (not shown) of the shutter opening and closing means 306. The private display control block 318 receives the monitor information from the monitor information obtaining means 328, and the image data frame sequence generating means 322 and the shutter voltage sequence generating means 324 based on the authentication level and the display security level of the user. The masking image generating means 326 is controlled. The monitor information obtaining unit 328 reads information such as the resolution of the monitor 304, the refresh cycle time, the vertical sync, the horizontal sync, and the like.

The image data frame sequence generating means 322, the shutter voltage sequence generating means 324, and the masking image generating means 326 correspond to the corresponding image data frame sequence and shutter according to the user's authentication level and display security level and the user's additional selection. A voltage sequence and a masking image are generated respectively. The shutter voltage sequence generating means 324 generates a shutter opening and closing sequence in synchronization with the image data frame sequence, and generates a voltage sequence corresponding to the shutter opening and closing sequence.

The dedicated driver 310 provides the masking image generated by the masking image generating means 326 to the video memory 328 according to the generated image data frame sequence, or the masking image by itself is instructed by the masking image generating means 326. Is generated and provided to the video memory 328, or the color table is controlled in real time. The dedicated driver 310 also controls the image transmission to the monitor 304 by causing the video controller 312 to switch the private image memory block and the masking image memory block according to the generated image sequence.

The transceiver 308 transmits a shutter opening / closing sequence or a shutter voltage sequence to the shutter opening / closing means 306. The transceiver 308 may also transmit an encrypted shutter voltage sequence to an authorized user using encryption means (not shown). The transceivers 308 and 310 may be implemented as a wired link such as USB or a serial link or a wireless link such as IR or RF (FM, AM, Bluetooth). The video controller 312, such as a graphics card, includes a video memory 328 and monitors the original private image received from the graphics driver 314 and the masked image received from the dedicated driver 310 according to the image data frame sequence. Display on the screen.

As shown in FIG. 3, the shutter opening / closing means 306 may include a transceiver 310, a decoder / authentication means 330, a shutter controller 332, and a shutter 334. The transceiver 310 receives an encrypted shutter opening / closing signal transmitted from the transceiver 308 and transmits it to the decoder / authentication means 330. The decoder / authentication means 330 decodes the shutter open / close signal to generate a shutter voltage sequence, and the shutter controller 332 opens or closes the shutter 334 completely or in an intermediate state according to the shutter voltage sequence.

The display security level is set to a performance level according to 'eye security' for unlicensed people who do not have shutters, and 'spy security performance' for non-licensed users who have different shutters. In general, the higher the level of display security, the lower the user's visual perception performance, such as user's visual comfort and image clarity. These display security levels can be defined in various ways. As an example, the first level is that an unauthorized person may not even perceive the approximate type of user private image even after viewing the monitor for a long time over a certain time. As the strictest privacy protection, for example, you don't know whether the user is doing a word or watching a video. The second level may be to perceive the approximate type of user image if the unauthorized user views the monitor for a certain time. However, some of the contents of the user image information cannot be grasped. For example, even if a user knows whether to watch a video, it does not know whether it is a movie or a chat. The third level can roughly grasp a portion of the user image information content if the unauthorized user sees the monitor for a certain time. However, most of the contents of the user image information cannot be grasped. For example, you cannot know the content of the word you type. You can tell if the user is watching a movie movie, but not the content. The fourth level can accurately grasp a part of the contents of the user image information if the unauthorized person sees the monitor for a specific time or more. However, most of the contents of the user image information cannot be grasped. For example, you might know a bit of the word you type. You can see a little bit of the content of the movie that the user perceives. The fifth level allows the unauthorized person to grasp a substantial part of the content of the user image information. However, I feel uncomfortable with my visual perception. As another example, an additional figure of merit may be added to this performance level to the extent that user private images and intentionally disturbing masking images can be recognized by an unauthorized person. In this case, various display security levels can be set, such as the performance level.

4 is a block diagram of a video controller 312 shown in FIG. As shown, the private video memory block is represented by P1 (Pm, and the masked video memory block is represented by M1 to Mn. P1 to Pm are frames operated by the system, and M1 to Mn are kernels for masking. These are masking frames generated by a dedicated driver 310, which is a driver, each memory block is selected through a method such as flipping, and a color table 402 serving as an image value (RGB value) converter. ) And block 404 to the monitor 304. As an example, each of the memory blocks P1 to Pm and M1 to Mm corresponds to one monitor frame size. Each time the CPU has a spare time, the image data is updated in the masking image memory blocks M1 to Mn, which are arbitrarily set in the video controller 312. 310 checks the vertical synchronization state Vsync_ST in a polling manner from the vertical synchronization status register 406 of the video controller 312 or the vertical synchronization interrupt generated by the vertical synchronization signal generation circuit 408 to the system. By hooking (Vsync_INT), one of the video frames P1 to Pm and M1 to Mn is selected in synchronization with the vertical synchronization signal, and the start address of the frame is set to the frame start address register of the video controller 312. (Not shown) so that the selected image frame is displayed.

In addition, the selected video frame may be displayed in synchronization with a specific horizontal synchronization signal. In the method of changing the color table, which will be described later, the private driver may change the color table arbitrarily and convert an arbitrary image value while the private image block is transmitted to the monitor, thereby converting the private image into a masking image. In addition, a filter driver (not shown) may be inserted instead of the dedicated driver 310 of FIG. 3. The filter driver functions like the dedicated driver 310 and works like a filter when the graphics driver 314 accesses the video memory 328. Alternate driver methods can also be used. This method integrates a dedicated driver function into an existing graphics driver and controls it as a driver.

In addition, as shown in FIG. 5, main functions among dedicated driver functions may be implemented at an application level. In this case, the dedicated driver 508 is responsible for transmitting the shutter opening / closing sequence to the shutter opening / closing means 306 through the transmitting / receiving unit 510 and controlling the color table in real time. 514 provides or masks the masking image generated by the masking image generating means 524 to the video memory 518 through the graphic driver 520 according to the image data frame sequence generated by the image data frame sequence generating means 322. According to the instruction of the image generating means 524, a masking image is generated by itself and provided to the video memory 518. The image control means 514 at the application level also causes the video controller 516 to control the transmission of the image to the monitor 504 by switching the private image and the masked image according to the generated image data frame sequence. The image control means 514 at the application level may be implemented by including a page flipping method using, for example, DirectX. In addition, the image control means 514 may create a virtual desktop and display it to the user. The image control means 514 can also extend the desktop and control the origin movement of the desktop.

6 is a block diagram of a wired / wireless communication interface between the main bodies 302 and 502 and the shutter opening and closing means 306 shown in FIGS. 3 and 5. The communication interface may be implemented as a wired link such as USB, IEEE 1394, serial link, or a wireless link such as IR, RF (FM, AM, Bluetooth). For example, in the USB wired link as shown in FIG. 6A, the USB system receives a signal indicating whether the shutter is opened or closed from the USB port 602 of the PC. The USB system receives + 5V power from the USB port 602 of the PC and boosts the voltage to + 12V, + 15V, etc. in the power module 606 to obtain a voltage required for shutter control. The shutter controller 608 receives a control signal from the USB CPU 610 and applies a voltage to the shutter 612 of the glasses. The control signal may be implemented by a modulation scheme such as Pulse Width Modulation (hereinafter referred to as "PWM") or a simple DC voltage. The wired link embodiment in which the decoder / authentication unit 632 is added to the shutter unit 626 is configured as shown in FIG. 6B. In the embodiment of FIG. 6A, the shutter controller 608 is included in the transmitter 604. However, in the embodiment of FIG. 6B in which the private image can be viewed only by an authorized shutter, the shutter controller 634 is a decoder / authentication means 632. ) Is included in the receiving unit 626, and the encrypted shutter opening and closing signal is transmitted to the receiving unit 626. In another embodiment, in a wireless link such as FIG. 6C, the transmitter 646 including the transmitter 652 is connected to a USB system of a computer, and the receiver 648 is a shutter 660, a receiver / decoder / authentication means 654. ), A power module 656, and a shutter controller 658.

7 is a system configuration diagram of a hardware external type embodiment according to the present invention. Compared to the SW method using software that implements most functions except the shutter unit and the transmission / reception interface unit, the hardware (hardware: HW) method refers to sequence control, generation of masking images, integration of private images and masking images, etc. It means a way to implement some core functions in hardware. In the HW method, there are various embodiments such as an external type, an independent card type, a graphic card embedded type, a graphic chip embedded type, and a monitor embedded type. These embodiments have in common that circuitry implements some key functions, such as the generation of masked images and the integration of private and masked images. Between these embodiments, such key functional circuits can be implemented with simple circuit changes, depending on where they are placed relative to the monitor 304 and the computer 702. The external type implements the core function circuit outside the monitor 304 and the computer 702, the independent card type implements the card type to be plugged into the computer, and the integrated graphic card adds the core function circuit to the graphic card circuit. That's the way it is. In addition, the integrated graphics chip is a method of adding the core functional circuit to the graphics chip, and the integrated monitor is a method of adding the core functional circuit to the driving circuit inside the monitor. Among the various forms, a description will be given on behalf of the HW type external embodiment shown in FIG. 7.

The HW scheme allows the video controller 710 to be connected via the core function circuit 704 rather than directly to the monitor 304. In the SW method, the dedicated driver performs the sequence control and masking image generation transmission function and the shutter opening and closing means connection function. In the HW method, the dedicated driver 714 is only responsible for the software and hardware connection function. In the HW method, sequence control, masking image generation, and integration of a private image and a masking image are implemented as one external circuit 704. That is, the core function circuit 704 includes a sequence controller 732 for generating a video sequence signal of a private video and a masking video, a transceiver 738 for transmitting the video sequence signal to the shutter opening and closing means 306, and a masking video. A masking image generating means 734 for generating a mark and a masked image transmitted from the masking image generating means 734 and a private image transmitted from the video controller 710 according to the image sequence signal to the monitor 304 as a public image. The monitor interface 736 is provided.

8 is a block diagram of an HW type external controller. The external controller 802 having the core function circuit is provided with control information (drive, stop, etc.) or sequence information from the computer's COM, PS2 or USB port 803, and informs the computer of the state of the external controller 802. . In addition, the external controller 802 obtains RGB, DVI signals, vertical sync signals (Vsync), horizontal sync signals (Hsync), and the like from the VGA output terminal 804 of the computer. The vertical synchronization signal Vsync or the horizontal synchronization signal Hsync generates an interrupt to the microprocessor 806 of the external controller 802 and the microprocessor 806 is a signal to be output from a computer or by applying a self-generated sequence. Set. At this time, the external controller 802 applies a PWM signal having an appropriate DC voltage or a duty corresponding thereto to the shutter opening / closing means 306 by using the transmission / reception unit 808 to determine whether the door is open or closed.

The masking video signal generator 812 manipulates the RGB and DVI signals of the private video obtained from the VGA output terminal 804 or generates random disturbance signals by itself. Masking image generation in HW method includes reverse phase generation using inverting circuit for RGB, DVI signal of original private image, circular induction image generation using differentiator or integrator circuit or various filter circuits, random noise masking image generation using random noise generator Implemented by the self-image storage module to create a disturbing image using the same, and also implemented by generating a combination of these methods as appropriate. The masking image signal generator 812 serves as a converter of the image value (RGB value). The method of generating a reversed phase inverts the signal based on an arbitrary reference value between the maximum value and the minimum value of the RGB signal to produce an analog signal that is reversed. In consideration of human visual perception characteristics, the reference value is arbitrarily adjusted to allow bias from the mean value. The blank signal source 814 may generate a gray frame by inputting a constant voltage to the monitor 304 irrespective of the RGB signal, and may be used for the purpose of smooth screen switching and the like. In addition, when the masking image is transmitted to the monitor 304, a method of making the masking image brighter than the private image by using a booster circuit is also used. The HW method can realize boosting, random noise generation, etc. without burdening calculation time.

9 is a system configuration diagram of a completely independent HW type external embodiment according to the present invention. Fully standalone HW is a standalone hardware that implements most of the functionality for private displays. This approach allows the target computer system 902 to connect via independent hardware 904 rather than directly connecting the video controller 710 and the monitor 304. Since most or all of the functionality for private display is implemented on independent hardware 904, there is less need to change the software portion for private display implementation to match the target computer system 902.

10 is a system configuration diagram of the SW / HW integrated external embodiment according to the present invention. This system implements a private display by integrating the SW method and the HW method. In particular, the dedicated driver 1004 generates a masking image by software using the masking image generating means 524, and the masking image generating means 734. ), The main focus is to create and integrate with hardware. In one embodiment, a basic masking image is generated by a software method, and then random noise is loaded or calculus effect and filtering effect is applied to the HW method.

11 illustrates a method of generating reversed phase using a nonlinear inverter and adding random noise. In FIG. 11, 1102 is a nonlinear inverter, 1104 is a random signal generator, and 1106 is a signal synthesizer. When the masking image is generated as shown, since the time required can be considerably reduced as compared with the case of generating the masking image by software as in the related art, the masking image can be generated in real time.

[Detailed division of video]

Masking images may be classified according to masking characteristics. Masking properties are indicated in superscript as follows: The original guided image is M ⁱ , the intentionally disturbed image is M ^d , and the connected image frame for smooth shutter opening or switching is smoothly expressed as M ^b . The masking image mixed with the original induced image and the disturbing image is M ⁱ when the original induced image is mainly added and the disturbing image is added to it, M ^d when the disturbed image is added and the original induced image is added to it. When the importance of the image and the disturbance image are equal or uncertain, indicate either M ⁱ or M ^d .

A circular induction image is a masking image calculated from a private image that is a circular image, and includes a reverse image, an inverse image of a reverse image or a reverse image, a filtered image of a circular image or a reverse phase, a shift image of a circular image, and the like. The disturbed image is a generic name of a masking image except the original induced image. A connection image frame for smooth shutter opening and closing or smooth image frame switching is an image frame inserted for smooth shutter opening and closing, and refers to a connection image frame such as a blank frame or a uniform gray frame.

In the private display where the frames of the private image and the masked image are continuously replaced, the human visual perception characteristics must deal with the temporal and spatial frequency characteristics, time integration, and temporal inhibition. Among the human visual perception characteristics, the temporal and spatial frequency characteristics are represented by the spatial contrast sensitivity function (hereinafter referred to as "SCSF") shown in FIG. 12A and the temporal contrast sensitivity function shown in FIG. 12B. : Hereinafter referred to as "TCSF". 12A is a spatial versus sensitivity function for luminance (○) and color (□), and FIG. 12B is a sensitivity versus time function for luminance (○) and color (○). As shown, the visual perception is sensitive to both luminance and color in the low spatial frequency and low temporal frequency domains, but is particularly sensitive to color. The visual perception is sensitive to both luminance and color in the low spatial frequency and intermediate time frequency domains, the intermediate spatial frequency and low time frequency domains, and the intermediate spatial frequency and intermediate time frequency domains. In other high spatial and high time frequency domains, the viewing angle is sensitive to luminance but insensitive to color. The critical flicker frequency (hereinafter referred to as "CFF") is the maximum responsive time frequency of about 55 Hz.

On the other hand, the physiology of the eye shows that the magnocellular pathway (referred to as "M pathway"), the parvocellular pathway (referred to as "P pathway") from the retina to the brain, Optic nerve signal pathways such as the K pathway exist. Table 1 shows the characteristics of the M and P paths, which occupy most of the paths. The temporal resolution is best for cones connected to the M path just around the fovea. The temporal resolution of cone cells connected to the povia P-path is medium and the time resolution of rods far from the povia is the worst. The rods have the best temporal sensivity.

The human eye is visually perceptual by time-integrated sampling. Conventional mechanical signal sampling only picks up signals at the time of sampling, while time integration samples the integral of the signal within the time integration period. The time integration section is adaptively variable according to stimuli intensity, stimuli size, background intensity, phase coherence, contrast, and the like. Basically, time integration takes place primarily in the photoreceptor. Thereafter, each optic nerve path leading to the brain has a different integral interval, and finally, information integration occurs in the brain. These multiple time integrations are aggregated so that time integration occurs during the adaptively variable time integration period.

The receptive field of optic nerve cells consists of a receptor that receives stimulus input. The center-surround receptive field structure of ganglion cells and LGN cells receives excitatory input from the center receptive field and from the surrounding receptive field. Receive inhibitor input. The center-peripheral accommodating field structure of these cells has center-surround antagonism with respect to luminance but not with respect to color (see FIG. 13B). Table 2 shows the structure of P cells in the P pathway. Here there is a slight time delay between the central response and the peripheral response (see FIG. 13A). At low time frequencies, since the time delay is smaller than the period of the response signal, it antagonizes the luminance and synergizes the color (see Fig. 13C). As the time frequency increases and the period of the response signal continues to decrease, a 180-degree phase shift occurs between the center response and the peripheral response, causing synergism with luminance and antagonism with color (see FIG. 13D). Due to this physiological time-suppressing structure, the optic nerve is insensitive to changes in luminance at low time frequencies and sensitive to changes in luminance of 5-30 Hz, as shown in FIG. 12, and is insensitive to changes in colors at low time frequencies and insensitive to color changes of 20 Hz or more. Done.

[Inverse generation]

As a masking image for masking the private image, an inverse image using time integration among the original induction images of the private image may be generated. If a private image and a reverse image are provided within the time integral section, a person perceives the average image with the reverse image because the person integrates the image during the time integral section.

The simplest possible inverse image is a static inverse image, where the static inverse pixel image value in the luminance-color space is the maximum pixel image value minus the original pixel image value (where the original image represents a private image). . For example, when one pixel image value of the original in the RGB space is (R ', G', B ') and the maximum pixel image value is (R' _m , G ' _m , B' _m ) is a _{(R 'i, G' i} , B 'i) = (R' m -R ', G' m -G ', B' m -B '). However, in a dynamic situation in which the original and reverse frames are constantly replaced, such as a closed display, the static inverse image cannot be a true inverse image due to the complex visual perception of the eye, and a dynamic inverse image must be obtained.

Dynamic inversion is calculated based on the perceptual domain. The general structure of image generation and visual perception is shown in FIG. 14. Images may be generated inside a computer through an image editing program or an operator's input, or an object in the light energy domain may be captured by a camera to have a value Ga in the computer domain. In general, the value Ga is gamma corrected with respect to the intensity or luminance of the light energy region. Color and luminance values of the light energy region are mainly expressed as tristimulus values, such as CIE XYZ or RGB space. The image value of the computer region corresponding to the RGB space is gamma corrected Ga = (R ', G', B '). The image value Ga in the computer domain is converted to the monitor input voltage and output and displayed by the monitor, and this process is expressed as the value I in the light energy region. The transfer function of this process is called the monitor transfer function f _rm , and the relationship of I = f _rm (G _a ) is established. The monitor transfer function f _rm is characteristically as shown in FIG. 15. Normally f _rm is modeled relatively accurately with a power function such as I = G _a ^rm . The intensity or luminance of the light energy region is perceived by the visual perception transfer function f _a between the luminance of the person and the perceived brightness, and a relationship of V = f _a (I) is established. The visual perceptual transfer function f _a is shown in FIG. 16. Normally f _a is modeled relatively accurately with a power function such as V = I ^a . As such, when the monitor transfer function and the visual perceptual transfer function are modeled as power functions And α are almost in inverse relationship. Accordingly, G _a and V are almost linearly proportional.

However, such a relationship is a result obtained in a static situation, and in a dynamic situation in which pixel image values change with time, the relationship is out of a linear relationship. Since the human visual perception is different depending on the temporal and spatial frequencies of the pixel image values, the linear proportional relationship between G _a and V is broken. When the image is reversed as a masking image in a private display, the nonlinearity of G _a and V according to the time frequency becomes important because the pixel image value is changed to high contrast with the time frequency of a specific region. Here, the time frequency is greatly affected by the refresh rate of the monitor and the frame sequence of the private and masked images. In conclusion, in order to obtain a reversed phase in a private display, a dynamic reversed phase should be obtained by considering the time frequency of the image together with the time integration.

As can be seen from the luminance TCSF curve of visual perception (FIG. 12B), the sensitivity drop at low frequencies and the increase in sensitivity at intermediate frequencies (5-30 Hz) are mainly due to time suppression. Because. This curve and related experimental results show that flashes of 5 to 30 Hz are perceived brightest (Bruke-Bartley effect). When the brightness change frequency of the flash exceeds CFF (approximately 55 Hz), time suppression is not affected much and is perceived as an average image by time integration.

The optimal dynamic reverse phase depends on the refresh rate. If the refresh rate is greater than 2CFF (approximately 110 Hz) and approximately half of the inverse phase is reversed between private images, the average image (composite image) is not visible to the naked eye because approximately half of the private image and inverse phase exist during the adaptive variable time integration period. see. If you give a reversed phase, the bright spots in the private image will be dark in the reverse phase and vice versa, so the frequency near (refresh rate) / 2 in addition to the refresh rate becomes the main frequency of the image. Therefore, if the refresh rate is 120Hz, P and M image replacement occurs around 60Hz, so the P and M image replacement frequency is larger than CFF, that is, (refresh rate) / 2 is larger than CFF, so the time suppression has little influence. It is perceived as an average image by time integration. Accordingly, when the refresh rate is greater than 2CFF (about 110 Hz), the dynamic reverse phase is almost indistinguishable from calculating based on the perceptual domain and calculating based on the light energy region. Suppose that the time integration interval is Δt _s , one pixel displays the private image with the intensity of I _p (t) for Δt _p and the masking image with the inverse phase for Δt _i with the intensity of I _i (t). . Here, Δt _s = Δt _p + Δt _i , Δt _p = Δt _i . In general, since the intensity response of the CRT and LCD monitor pixels is shown in FIG. 17, I _p (t) and I _i (t) are variables. For convenience, the ideal monitor calculates the energy equivalent of the actual monitor and depends on the nature of the time integration. An ideal monitor displays the intensity corresponding to the image value Ga in the computer domain in square wave form. Strength of these ideal monitors , With the intensity representative value, , Calculate , . This yields an ideal monitor that is energy equivalent to the actual monitor. One pixel like this Private video Marked with the strength of Reversed masking footage while Approximate that it is expressed by the intensity of. In fact, this approximation is valid if the product of intensity and time is the same within the time integration period, because it is perceived as the same image regardless of the waveform.

Average video intensity of closed and reversed images Can be obtained simply as in Equation 1. This is represented by the image value of the computer domain, and is represented by Equation 2 below. Where d is an average image, p is a private image, and i is an inverse image.

First, find the average image. The maximum and minimum values of the current and If it is set to the average image value from the following equation (4) To calculate. Next, private video value With respect to the inverse value can be simply obtained from the following equation (5) or (6).

On the other hand, the static inverse phase is obtained as shown in Equation 7, which is not appropriate in a dynamic display situation. For example, using RGB space and 24-bit color in the computer domain = 255, = 0 Gamma of the monitor Is 2.2, the average image value is ≒ 186, The inverse of = 170 is = 201. In the case of static reverse phase, the average image value ≒ 128 The inverse of = 170 is = 85. In RGB space, each RGB component is calculated in this way.

If the refresh rate is less than 2CFF (approximately 110Hz) and approximately half of the inverse phase is reversed from the private image, the P / M image replacement frequency is smaller than CFF, resulting in incomplete time integration and time suppression affecting the tactics. Dynamic reversed phase calculations, as in one, are not accurate. The inverse phase should be calculated taking into account the effects of incomplete time integration and time suppression, and taking into account the modeling uncertainties of the monitor and visual perceptual transfer functions. However, the calculation of inverse analytically is limited because the human visual perception process is not yet known precisely.

Therefore, in this situation, the dynamic reverse phase is calculated using the following dynamic pattern experiment method. First, we need to set the maximum and minimum and If it is set to the hatched portion in the pattern as shown in Figure 18a , And other white squares And make this pattern original and reverse the pattern , Other white squares ), Half by half to display. When viewing this image, the higher the sharpness of the T, the more incomplete the time integration occurs. If the T is not perceived, the time integration is complete. This image captures the degree of time integration under the current refresh rate. Next, the approximate average image value is calculated by using Equation 8 below.

Next, the approximate average image value of the hatched portion of FIG. And other white square part To make this pattern circular, and reverse the pattern , Other white squares ), Half by half to display. Average image value The image value with the lowest T sharpness is displayed closest to the average image value. This value is found and selected as the average image value. Next, each original value Dynamic Inverse with respect to Value It is a step to obtain. Average image value obtained using the steps as described above Approximate dynamic inverse value from Calculate

Approximate average image value of hatched portion of FIG. 18A And other white squares To make this pattern circular, and reverse the pattern , Other white squares ), Half by half to display. Inverse value The image value with the lowest T sharpness is displayed closest to the dynamic inverse value. Given this value Dynamic Inverse for To be selected. In this way all Dynamic reversed phase values are selected and stored for each value (R, G, B in the case of RGB space). In the above, the dynamic pattern experiment method using the pattern of FIG. 18A is taken as an example, but the above procedure may be tested using various patterns including FIG. 18B.

The dynamic pattern test method described above is a monitor in which the ratio of provision of private images and reversed images is out of 1: 1, the refresh rate is lower than 2CFF (approximately 110 Hz), or the monitor's fluorescent material is degraded or the monitor's actual transfer function is modeled. This is particularly effective when moving away from the transfer function. In addition, the refresh rate is higher than 2CFF (approximately 110Hz), and the effect of time suppression remains slightly even when the private image and reverse phase are replaced by about half, and the actual transfer function of the monitor may deviate from the modeled monitor transfer function. Using the dynamic pattern experiment method, a more accurate dynamic inverse can be obtained.

The private display software stores in advance the dynamic inverse value calculated by the calculation method or the dynamic pattern experiment method for each refresh rate, the ratio of the display of the inverse image and the inverse image, and the monitor type (CRT, LCD). This allows the user to use the appropriate dynamic inverse value depending on the usage environment or choice. In addition, since the characteristics are slightly different for each actual monitor, the private display software may be provided including the dynamic pattern experiment method software so that a user may calculate and use the dynamic reversed value directly.

[How to change the color table]

When the reverse phase is generated as the masked image, the reverse phase may be generated by calculating and storing the reverse phase value corresponding to the pixel image value for each pixel of the private image in the video memory. This method takes a lot of computation time because the pixel inverse value calculated after each software calculation is stored in memory. This method was difficult to generate in real time reverse phase, so it could be applied only to the private display of still images. In order to generate a real-time induced image such as reverse phase in real time, the present invention uses an image value conversion method using an image value (RGB value) converter. As a preferred embodiment of the present invention, a color table included in a video controller such as a graphics card is used as an image value converter to generate a real-time original guidance image by arbitrarily changing the color table according to a conversion rule. In the process of transmitting the image value of each pixel in the video memory to the monitor, a typical computer system converts the image value by referring to the color table when sending the image value Ga of the computer domain to the display, and then converts the converted image value to D / D. A conversion can be sent in analog format such as RGB voltage or digital format such as DVI.

The original color table is intended for fine correction of the image, and generally there is little difference between the image value Ga and the converted image value. As an example, when the image value Ga is (R, G, B) as shown in FIG. 19, the converted image values R_out, G_out, and B_out output the image value Ga almost as it is and require fine correction of the image (Fig. In 19, only R_out = 187) is slightly changed when R = 188.

In the present invention, the conversion value of the color table is changed to the image value of the computer area by an arbitrary conversion rule. A first color table for the private image and a second color table for the masking image are prepared to change the actual color table of the video controller to the value of the corresponding first or second color table when each image is displayed. In this case, a different image is generated according to the changed color table converted value, and a difference between the image value Ga and the converted image value may be large, and may have opposite characteristics. As an example, if each Ga-corresponding converted image value corresponding to the dynamic inverse phase is inserted into the color table instead of the original color table, the dynamic inverse image is displayed even if the original image is displayed as it is. 20 illustrates an embodiment in which a dynamic reverse phase is implemented by a color table changing method using R = 187 as an average image value. At this time, the pixel image value Ga = (R, G, B) is R = 0-> R_out = 255, R = 255-> R_out = 0, R = 186-> R_out = 188, R = 187-> R_out = 187 , R = 188-> R_out = 186 is output as a converted video value according to the changed conversion rule. The color table change method generates an inverted phase by changing only a few color table values without having to calculate an inverted value for each pixel image value.

When the color table is changed according to an arbitrary rule in the method of changing the color table, not only the dynamic inverse value but also an arbitrary converted image may be obtained according to the changed color table. In an embodiment, when the Ga-corresponding converted image value of the color table is set to a constant value for each Ga value, an image irrelevant to the color table input image may be obtained. In this way, a uniform gray image independent of the input image, a gray image with uniform color, and a uniform color image can be obtained.

In the method of calculating a circular induction image, such as generating a reverse phase by calculating an inverse phase value for each pixel of the private image, for example, the original induction image calculated from the private image of the P1 memory of FIG. 4 is stored in the M1 memory and the image of M1 is stored. Will be displayed on the monitor. In contrast, the color table changing method does not directly manipulate the video memory, but changes only the color table according to the calculation rules of the original induction image and outputs the private image of the P1 memory to the monitor to display the original induction image masking image. In another embodiment, when the private image of the P1 memory is stored in the M1 memory or the like and only the color table is changed and the image of the M1 memory is output to the monitor, the original induction masking image is displayed.

The driver program changes the color table conversion values to match the graphics card's vertical synchronization, to a specific horizontal synchronization, or asynchronously. The embodiment of changing the color table value in accordance with the vertical synchronization detects the vertical synchronization in a polling or interrupt manner in the driver and replaces the value of the color table with the new value as soon as the vertical synchronization occurs. In the case of the interrupt method, the existing graphics driver is hooked to change the color table value whenever an interrupt occurs. Changing the value of the color table takes tens of microseconds or less, so the effect of the changed color table is immediate. In another embodiment of the present invention, the real-time image value conversion is implemented by using an HW type inversion circuit, a calculus circuit, and the like as an image value converter.

[Intentional Disturbance Masking Image Generation]

The disturbance image in the present invention is defined as a generic term for a masking image excluding a primitive induction image, which has no dependency on a closed image. Conventionally, white flash images, random images, still images, and screensaver images have been used as disturbing images. However, this is an effective method only for masking a specific private image, which is a systematic and strategic disturbance image that can be used for general private images. It does not consider human visual perception.

The present invention considers human visual perception characteristics and proposes a systematic and strategic disturbance image generation and management method. 21 is a flowchart illustrating a disturbance image generation and management method according to an embodiment of the present invention. First, a disturbance image in consideration of human visual perception characteristics is produced and stored (2101). The disturbed image is produced using frequency characteristics, brightness contrast, color contrast, and cognitive content characteristics that are sensitive to visual perception. In one embodiment, a disturbance image is produced to effectively mask any private image that can be generated as one disturbance image. In another embodiment, a disturbing image that emphasizes specific characteristics is produced to better mask a specific private image. The disturbed image file may have a size corresponding to the monitor frame or may be a unit of an image block larger or smaller than the monitor frame. In one embodiment, several disturbing images are generated. In another embodiment, multiple disturbance images are produced for each display security level.

Next, if multiple disturbance images are produced, an index or file name is set based on the disturbance image characteristics for searching necessary for generating a suitable disturbance image (2102). Next, a private display mode is started by the user or the system (2103). In step 2103, it is determined whether the disturbed image is used as a masking image. If it is determined to use the disturbed image, it is selected whether to inspect the characteristic of the private image currently being displayed (2104). If no test is selected, go to step 2106. If you select the test, go to check the spatiotemporal frequency, contrast, and cognitive content characteristics of the closed image (2105). Content-based inspection is used as an embodiment of the present invention to examine the characteristics of a closed image. For example, the cognitive content characteristics of a private video are identified based on the nature of the most recently used window or the active window (whether a word, web browser, video player, etc. is used). As another example, the current application SW is examined to determine the cognitive content characteristics of the private video according to the typical content of the application. As another example, a user may input anticipated private image characteristic information or select a mode based on an application to be used in the future. In another embodiment, the frequency characteristic of the private image is detected in real time to generate a disturbing image corresponding thereto in a later step. To reduce the computing burden, only part of a private video can be frequency converted.

In step 2106, the disturbance image generation guideline is established. To determine the level of frequency, brightness, color, and cognitive content characteristics of a disturbing image to generate a suitable disturbing image, if the display security level is set, corresponding to the set security level, and if not, the security level of the disturbing image. Establish guidelines for The guideline is determined by a systematic methodology including a space-time frequency characteristic condition, brightness contrast and color contrast condition, and cognitive content characteristic condition of a disturbing image to be generated. The guidelines include such conditions as generating a disturbing image including a text image and how much color contrast to use. The systematic methodology for defining the guidelines is programmed and provided in the private display control block 318 or the masking image generating means 326 of the system. In this step, if the characteristic of the private image is identified through step 2105, a disturbance image generation guideline suitable for the private image characteristic is established, and if it is not determined, an effective effective disturbance image generation guideline is established.

Next, the private display control block 318 and the masking image generating means 326 of the system search for a disturbing image that satisfies the condition of the generation instruction (2107). As another example, the user may directly search through the user interface 320. When a suitable disturbance image is found, the selected disturbance image is generated and loaded into the memory (2108). If there is a disturbing image file in units of image blocks smaller than the monitor frame, a plurality of disturbing image blocks may be loaded to configure one disturbing image frame.

When the initial image is loaded into the memory, the disturbed image is updated in time (2109). The temporal updating method may include, for example, loading and updating a new image that may occupy the entire monitor frame into a memory, or in another example, moving or rearranging one or more disturbing image parts or image blocks constituting the monitor frame in time, or another example. Load a new image block file into memory to update a part of the image, or move the image by randomly changing the start address of the memory, which is the flipping address when flipping another example, or by a real-time generation program selected by the system It uses the method of updating the image through. The time frequency of the image generated when the image is updated in time is set according to the disturbance image generation guideline, and the image is temporally updated by a time varying algorithm based on the guideline. Human visual perception characteristics are considered to establish the conditions of spatio-temporal frequency, brightness and color, and cognitive content characteristics of the disturbed image used in the disturbed image generation method.

22 shows spatial and temporal frequency characteristics of human visual perception and typical image data. Where tf1, tf2, tf3, sf1, sf2, and sf3 are frequencies that distinguish low frequency, intermediate frequency, high frequency, and limit high frequency, respectively. It has a value of ~ 15cycles / deg and sf3 ~ 40cycles / deg. First, FIG. 22A shows physiological evidence of human visual perception as a response frequency region of P cells and M cells, which are representative optic neurons. FIG. 22B illustrates visual perception characteristics obtained by integrating the SCSF of FIG. 11 and the TCSF of FIG. 12. Human visual perception cannot detect high frequency components of an image exceeding the time limit high frequency tf3 or the space limit high frequency sf3. As shown, in the low spatial frequency and low temporal frequency regions, the visual perception is sensitive to both luminance and color, but in particular the visual perception is sensitive to color. The visual perception is sensitive to both luminance and color in the low spatial frequency and intermediate time frequency domains, the intermediate spatial frequency and low time frequency domains, and the intermediate spatial frequency and intermediate time frequency domains. In other high spatial and high time frequency domains, the viewing angle is sensitive to luminance but insensitive to color.

22C shows the spatial frequency and temporal frequency characteristics of typical image data. In the present invention, the image is divided into a text image and a non-text image according to the cognitive content or meaning of the image. Text images include characters and symbols, and non-text images include photographed images, pictures, and graphic images except text. The photographed image is an image obtained by photographing means such as a camera, and includes a photograph, a movie, and a TV broadcast image. In the present invention, a still image and a moving image are defined according to time-frequency characteristics of an image. Still images are defined as images in which text and non-text images are fixed in time or almost still. A video is defined as an image in which text and non-text images move in time. As well as moving pictures in the normal sense such as computer graphics moving pictures, movies, and TV broadcasting video, as well as moving pictures over time, text is defined as moving pictures. The captured video, graphic video, and text video may have different spatiotemporal frequency characteristics. 22C illustrates frequency characteristics of a text image, a video, and the like in a conventional sense. Text images typically occupy high spatial frequencies and low temporal frequency domains from intermediate spatial frequencies. Still images, such as pictures and photographs, generally take up the low to high spatial and low temporal frequency ranges. Video occupies the widest frequency range.

In the present invention, the disturbance image is produced in consideration of the spatio-temporal frequency of luminance and color, and is generated and updated in consideration of the spatio-temporal frequency characteristics in the generation and update phase. Preferably, the disturbing masking image is generated to cover the temporal and spatial frequencies of the main luminance and color of the private image to be masked. In another embodiment, the disturbing image has all frequency components to which the visual perception responds. In another embodiment, a disturbing image in which more image components of a frequency sensitive to visual perception are inserted is generated. Since private videos have a wide variety of frequency components depending on the content of the video, if you create a disturbing video separately from the private video, you may not expect the private video, so some videos may be masked well but others may not mask. It is easy to occur. In one embodiment, in step 2105, the space-time frequency of the closed image is examined and determined. When the characteristics of the closed image are grasped, the disturbed image is preferably generated to cover the main temporal and spatial frequencies of the closed image and to insert more image components of frequencies sensitive to visual perception. For example, if a private image is a typical text image, the disturbance image is generated to cover a high spatial frequency and a low temporal frequency region from an intermediate spatial frequency. In addition, the disturbance image is generated to insert more image components of the frequency sensitive to visual perception than the private image. Using the above, the frequency condition of the disturbance image generation guideline suitable for the private image characteristics is established. In another embodiment, when the spatiotemporal frequency of the closed image is not known, a frequency condition including as many image components of a frequency band as sensitive to visual perception is established so that an effective disturbed image is generated on average.

Disturbance images should be clearly visible than private images. To this end, brightness and color contrast sensitivity play an important role, as well as space-time frequency sensitivity of visual perception. In an embodiment, each image element of the disturbed image has high luminance contrast or high color contrast. To systematically generate images with high brightness and color contrast sensitive to human visual perception, it is necessary to generate them in a quantified system so that the image contrast is proportional to visual perception sensitivity. In the present invention, as the quantified system, a color space composed of luminance and color axes is used. As is well known, there are several kinds of color spaces such as RGB, CMY, Yuv. However, many color spaces do not fully consider the human visual perception system and are not configured to be proportional to visual perception sensitivity. A color space proportional to human visual perception sensitivity is the CIE L ^* a ^* b ^* (simply CIE Lab) color space standardized by the Commission Internationale de I'Eclairage (CIE). 23 illustrates the CIE Lab color space. The color space consists of three independent visual perceptual cells of the human visual perceptual system, with the visual perception of the luminance cell, the red-green cell, and the yellow-blue cell in an orthogonal axis. In the present invention, a difference between two image values A and B in the CIE Lab color space is defined as a luminance color difference, and the larger the luminance color difference, the higher the contrast between the two image values. The vertical component of the luminance color difference is the luminance difference and the horizontal component is the color difference. The luminance difference is defined as the luminance contrast, and the color difference is defined as the color contrast.

The image value C point of FIG. 23 is a center point corresponding to the center value of the color space. If the monitor does not sufficiently display the luminance and color of the color space, or if the range of image values displayed on the monitor is limited through image value compression such as a color table change method, the center point may move. Preferably, in the present invention, the contrast of the disturbed image is determined based on a visual perception proportional color space such as a CIE Lab color space. In one embodiment, the disturbed image is generated to have an outer extreme image value of the color space on average compared to the private image. Predetermined conditions regarding luminance and color contrast are prepared, and a disturbance image is produced or generated based on the conditions. As an example, a condition is created such that the average luminance difference or average color difference from the reference value such as the color space center value to each image value of the disturbing image element is larger than a predetermined value. An image satisfying the above condition has an average image value outside of the color space. In this case, since the disturbed image value is biased at one extreme of the color space to produce an image with high contrast, as a condition for producing a high contrast image, the color difference between the average image value of the disturbed image element and the reference value is a predetermined value. The following conditions are provided. In fact, when producing or generating an image, one embodiment proceeds by evaluating whether it is a high-contrast image that satisfies a condition, and in another embodiment, evaluating whether the condition is satisfied approximately and qualitatively. Create An example of qualitative evaluation of a condition takes into account the area ratio of a pair of image elements that make up the image, such as an object / background. If the area difference between the object and the background yarn is large, the average image value moves toward the larger area, so the average contrast becomes smaller. In this case, the color difference between the average image value and the reference value exceeds a predetermined value and violates the condition. Disturbance images are generated by qualitative evaluation so that the area ratio of a pair of image elements does not make a big difference.

In accordance with an embodiment of the present invention, a method of producing and generating a disturbing image based on a visually proportional color space, such as a CIE Lab color space, selects an image value of a disturbing image element using a visually proportional color space. Convert to color space image values in the same computer domain. The conversion may be accurately calculated through calculation, or may be roughly and qualitatively converted. That is, even if a corresponding RGB color space image value is generated approximately by referring to the CIE Lab color space, an intended contrast image can be obtained.

According to the present invention, a disturbance image having a high contrast between at least one of brightness and color contrast between an object and a background of an image or an image pattern is generated. Preferably, a disturbing image having high contrast at various luminance and colors is generated. As one embodiment, if the characteristics of the private image can be grasped, a disturbing image having a higher contrast than the private image is generated on average. As an example, when the closed image is general text, since the luminance contrast is higher than the color contrast, a disturbing image having a higher luminance contrast is generated. Preferably, a disturbing image having one or more high color contrast image elements is generated in the high luminance contrast image element.

24 is a diagram illustrating a visual perception process. Perceptual grouping (combination) rather than images with no cognitive meaning makes images with specific cognitive meanings well perceived. When stimulus 2402 stimulates the retina, a preattentive process (2404), a fast visual perception process without feedback, is perceived. At this time, perception or perception of the form does not occur, and the presence or absence of a stimulus, change, or movement is perceived. The person then consciously or unconsciously focuses attention (2406) to perceive the shape and characteristics of the part that is currently focusing. In this case, feedback is received from the high level cognitive information 2408 such as memory and reasoning, which is used in the perception process, and the stimulus is continuously collected from the stimulus source. Next, through the centralized attention 2410, the perceived components 2412 are combined to make a comprehensive perception. In focused attention 2414, the perceptual grouping perceives images with cognitive meaning.

Perceptual grouping is the most important issue in terms of its relationship with high-level cognitive processes. According to the Law of organization of Psychophysics, perceptual grouping by simplicity, similarity, continuity, proximity, common movement, and meaning Becomes In addition, perceptual grouping is performed by time and space phase coherence. If a phase change occurs in any part of the pattern with one spatial frequency, that part is conspicuous. This is due to grouping by spatial phase coherence. If a pattern with one time frequency occurs at any point in the phase shift, the part is conspicuous. This is due to grouping by time phase coherence. The image is modulated in a frame sequence. When the image is presented to the user as a repeating replacement sequence and a phase change (non-repetitive sequence) of the sequence occurs, the image at that time becomes more visible and a minute shimmer is felt. Since the phase change in the time integration period is not felt by humans, the grouping by the time phase coherence is closely related to the time integration period of the visual perception. Since the image elements with perceptual grouping form a cognitive meaning, it is easy to be perceived as a whole even if parts of the elements are hidden or disturbed. Private videos are generally well grouped among themselves. Therefore, the masking image should be grouped with the private image so that the public image (the combined image of the private image and the masking image) is seen as an image having no cognitive meaning or an image having a different meaning. The masking image may perform a function of cutting off the grouping of the private image and a function of clearly grouping the masking images themselves so that the public image is displayed as an image having a different meaning from the private image.

In one embodiment, the disturbed image is composed of images having different temporal phase coherence and spatial phase coherence so as to cut off the temporal phase coherence and the spatial phase coherence of the private image. Originally induced images, including dynamic inversion, intermingle with private images to prevent grouping of private images.

To mask the text of the private image when the text image region of the private image is larger than a predetermined level, the disturbed image in the present invention has a text image element in at least one region. When the non-text image area of the private image is large, the disturbing image has a non-text image element in at least one region to mask the non-text of the private image.

If the edge of the image with cognitive content is clear, the visual angle is more sensitive, thus creating a disturbing image element having sharp edges. When producing a disturbing image including a captured image element, since a general captured image does not have a large contrast of the image and the edge is not clear, the embodiment further includes a process of artificially sharpening the edge. Images are processed and stored using image processing techniques such as histogram equalization, contrast enhancement, and homomorphic filtering. In another embodiment, the disturbing image includes an element having a three-dimensional stereoscopic effect. In another embodiment, the high-contrast pattern is a disturbing image including an image element causing dizziness or illusion, such as turning round and round.

In the present invention, by displaying two kinds of images while replacing them in order to obtain a noticeable disturbing image, it is tested which kind of image elements are prominent. Through such a test method, disturbed images are made from image elements judged to be easy to see. The disturbance image is made by combining images having sensitive frequency characteristics, contrast characteristics, and cognitive content characteristics as described above. Preferably, a disturbing image is produced by including an image element having a high luminance and color contrast, a frequency sensitive to visual perception, and having cognitively meaningful contents. For example, a high luminance contrast image component having an intermediate time frequency and an intermediate spatial frequency and a high color contrast image component having a low spatial frequency may be included in comparison with other frequency components. In order to mask patterns with low spatial frequencies in closed images, high color contrast is used and high luminance contrast is added. To mask typical text and high spatial frequency images, high brightness contrast is often used and color assimilation occurs, so color contrast images are added.

In an exemplary embodiment, different conditions may be set for each area of the monitor frame in the disturbance image generation guide establishment step 2106. Perturbation images that are more sensitive to visual perception than the other areas of the monitor are generated. For example, compared to the upper or lower area of the monitor, the center of the monitor generates a disturbing image that is more sensitive to visual perception. In this case, one disturbing image file may be loaded to configure a monitor frame, and two or more disturbing image blocks may be loaded to configure a monitor frame.

Since afterimages are relatively high in the lower part of the monitor, the partial disturbance images are generated as images that are less sensitive to visual perception than the center part. Since the upper part of the monitor has a slow shutter response, a user may perceive a little masking image, so in one embodiment, this area is also generated as a disturbing image which is less sensitive to visual perception.

In another embodiment of the present invention, a disturbance image is produced and generated by combining two or more different cognitive content characteristics, space-time frequency characteristics, and contrasting image elements. A spatial distribution arrangement and a space overlap arrangement are used as embodiments for spatially combining the above. In the spatial distribution arrangement, image elements having different characteristics are arranged for each monitor frame region of the disturbed image. As an example, images are produced by combining text images in one region and non-text images in another region according to different cognitive contents. As another example, an image having a high brightness contrast is blended in one region and an image having a high specific color contrast in another region. If a large area consists only of a disturbing image having one image characteristic, a situation in which masking is insufficient occurs. For example, if you have a large disturbing video area composed of text with black and white brightness contrast, masking will not work properly if a private picture such as a color picture is displayed in this area.

FIG. 25A illustrates an embodiment of producing and generating a disturbing image by arranging a space distribution. In the above, a text image and a picture and a picture image having luminance or color contrast are space-distributed. In the spatial overlapping arrangement, image elements having different characteristics are overlapped and arranged in one region. FIG. 25B illustrates an embodiment of producing and generating disturbed images by spatially overlapping arrangement. 25C is a diagram illustrating a process of constructing a disturbing image by using a spatial distribution arrangement and a spatial overlap arrangement. In actual operation, the element or part of the disturbing image is updated by moving or transforming with time or by loading a new image element.

In the present invention, the disturbing image preferably includes an image element having characteristics similar to the private image. In one embodiment, the disturbing image includes an image element having a different characteristic from an image element having characteristics similar to those of the private image. In general, image elements having similar characteristics are mainly used to stop perceptual grouping of private images, and image elements having different characteristics are mainly used to cause perceptual grouping different from private images. If the private picture is a text picture, the disturbing picture shall contain at least textual picture elements. If the private image is a photographic image, the disturbing image should contain at least non-text image elements.

In another embodiment, the masking image is displayed more frequently by making the disturbing masking image frame rate larger than the private image frame rate. In another embodiment, a computer domain image value (Ga value) compression method is used. This method compresses the range of Ga values (RGB values) of the original image (private image). For example, when the masking image has a range of, the range of the masking image is larger than that of the private image, just as the original image (the private image) has the range. This Ga value compression method has the effect of making the disturbing masking video frame rate equivalently larger than the private video frame rate. Preferably, Ga-value compression is implemented by color table compression.

In an embodiment of the present invention, the entire private display management server transmits an advertisement image to a private display user computer through a network, and uses the transmitted advertisement image as a disturbing image among masking images.

[P / M image sequence and shutter opening / closing sequence generation process]

After the user is authenticated and selected the display security performance level, the P / M image sequence and the shutter opening and closing sequence are generated according to the user authentication level and the display security performance level. 26 illustrates a process of generating a P / M image sequence and a shutter opening / closing sequence. After authenticating the user, the user selects from a preset display security performance level (2602). Next, a P / M image blending ratio and a blending rule satisfying the display security performance level and user visual perception performance selected by the user are selected (2604). Next, a P / M image sequence is generated according to the blending rule (2606). Next, a shutter opening / closing sequence that satisfies the selected security performance level and user visual perception performance and corresponds to the P / M image sequence is generated (2608). Next, in step 2610, it is determined whether a satisfactory shutter opening / closing sequence has been generated, and ends if a satisfactory shutter opening / closing sequence has been generated, and goes to step 2612. Step 2612 selects whether to reselect the formulation rule. If no re-selection is selected, go to step 2606 to create a P / M image sequence and continue to the next step; if reselect is selected, go to step 2604 to reselect the P / M image blending ratios and blending rules and follow up. Continue with the steps.

[Mixing ratio and formulation rule]

In step 2604 of FIG. 26, a P / M image blending ratio and a blending rule are selected. In this case, the current refresh rate of the monitor, the response time of the monitor pixels, and the optical response characteristics of the shutter are all taken into consideration. In addition, attention should be paid to the characteristics of the P / M image to be blended (reverse or disturbing image, etc.). The conventional P / M image sequencing method is not effective in various performance aspects because the above four factors are not considered when generating a sequence. In addition, the shutter opening and closing sequence should be considered when generating the P / M image sequence, and conversely, the P / M image sequence should be considered when generating the shutter opening and closing sequence.

Conventional P / M image combining has been to combine masking images having the same characteristics with respect to closed images. That is, masking images having the same characteristics were combined in such a manner that all of the masking images were reversed or all random images, and P / M image sequences were generated under such a mixing rule. In the P / M image blending ratio and blending rule of the present invention, images having different characteristics may be blended.

In one embodiment, a masking image in which the original induced image and the disturbing image are mixed is combined with a private image. The mixing of the original guided image and the disturbed image is a method of mixing image components having different characteristics such as the original guided image and the disturbed image in one masking image frame, and the masking frame M ⁱ mainly composed of the original guided image and the disturbed image are added. Or a dominant masking frame M ^d and combine them into a private picture and frame to frame. In this case, the masking image M ⁱ is a pure circular induction image or a summation image in which the original induction image is dominant compared to the disturbing image, and the masking image M ^d is a summation of the original induction image and the disturbance image, which is mainly a disturbing image or a certain ratio of the disturbing image. Indicates. The average compounding ratio can be variously determined as P: M = 5: 5, P: M = 4: 6, and P: M ⁱ : M ^d = 4: 4: 2. In the case of combining the P: M ⁱ : M ^d method, M ⁱ may be a mixed image in which M ⁱ is a random image having a constant amplitude in reverse, and M ^d may be an image frame having characteristics of a disturbing image. have. In another embodiment, the combination video frame M ^b is combined with P: M ⁱ : M ^b and P: M ^b : M ^d . For example, it may be blended in the same ratio as P: M ⁱ : M ^d : M ^b = 4: 4: 1.5: 0.5 (in this case, P: M = 4: 6). In another embodiment, two types of disturbance images may be combined, such as a disturbance image that is mixed with a private image to prevent grouping of the private image and a disturbing image that is grouped with each other.

Using the color table changing method, it is convenient to generate a masking image mixed with the original induced image and the disturbing image in real time. In the method of mixing the original induced image and the disturbed image in one masking image frame, in one embodiment, the mixed image of the original induced image and the disturbed image may be generated using a color table changing method. In another embodiment, a mixed image may be generated by generating a primitive induced image by using a color table changing method and adding a disturbance image by calculating pixel image values of some pixels. In another embodiment, a private image is copied to a masking image memory area and a disturbance image is added thereto to generate a combined image of the private image and the disturbance image. When the image is sent to the monitor, a color table change method is used to transmit the inductive conversion image of the image so that the original image and the disturbance image are generated and transmitted to the monitor. In addition to creating masking frames M ⁱ and M ^d and combining them between private images and frame-to-frame, the color table change method is convenient to create in real time.

In one embodiment, when combining P: M ⁱ : M ^d , M ⁱ is generated in real time from a private image using a color table change method, and when M ^d is updated relatively slowly by pixel image value calculation, the overall real time is obtained. Mixed masking images can be provided. In another embodiment, when combining P: M ⁱ : M ^d , M ⁱ is rapidly generated in real time using a color table change method, and M ^d is relatively slow, and the private image is stored in the masking image memory area several times per second. Create by copying and adding disturbance image to it. To transfer M ^d to the monitor, you can use the color table change method just like M ⁱ . Since M ⁱ can effectively mask a fast changing private image and M ^d can slowly mask a slowly changing private image, in one embodiment, the ratio of M ⁱ : M ^d is differently adjusted according to the characteristics of the private image. For example, if there are a lot of still images in a private video such as text work, increase the ratio of M ^d , and if there are many videos, increase the ratio of M ⁱ . The ratio adjustment may be automatically adjusted to a ratio set by the private display software to identify the characteristics of the current private image by the identified characteristics, or may be adjusted by the user. In addition, the ratio adjustment range is determined by limiting the display frequency of M ⁱ and M ^d per second in consideration of the current refresh rate of the monitor and the sensitive frequency characteristics of the human visual perception.

Since it is common to make the masked image more prominent than the private image, it is often necessary to reduce the contrast of the private image compared to the contrast of the masked image. When the contrast of the image is reduced, pixel image values can be calculated and reduced, but this takes a lot of calculation time. Reducing the contrast of an image by transformation is defined as color space dynamic range reduction. Here, dynamic range is a concept such as the maximum contrast of the color space (difference between the maximum value and the minimum value), the range of monitor brightness, and the range of voltage output to the monitor. According to an embodiment, the range of the monitor voltage may be reduced in hardware only when the private video is output in the HW method. In another embodiment, when the private image is output in the HW method, the private image may be output as it is, and when the masking image is output, the voltage of the monitor may be amplified. In the SW method, there is a color space dynamic range reduction using a color table changing method. For example, by changing the color table, the input RGB range [0,255] is reduced to the output RGB range [56,255].

In the private display, the user only sees the private image that is part of the monitor image, so the screen is darker than in the normal mode, so there is a need to increase the overall brightness. This can be accomplished by amplifying the voltage into the monitor and shifting the color table conversion upwards towards overall brightness. A 3M privacy filter or brightness enhancement film is used to improve the relative brightness of the monitor. In this process, user adjustments such as brightness adjustment button and contrast adjustment button on the monitor should be considered.

When you create a dynamic reverse image and swap it for a private video, you see a nearly uniform gray screen with bare eyes. However, there is a noticeable problem because a uniform gray screen is displayed as a background even though fine color and spatial patterns remain because the time integration is not perfect. This is especially true at medium to high (tf1 to tf3) time frequencies and at medium to high (sf1 to sf3) spatial frequencies. To solve this problem, as an example, a colored dynamic inverse image is provided. A color dynamic reversed image is an image in which the Ga value of a specific color is increased or decreased so that the color becomes slightly shaded gray when it is replaced with a private image. In one embodiment, changing the color of the dynamic reversed image that is colored by the color table changing method during one monitor frame time during which the masked image is displayed may generate multiple pieces of different colors in one masked image monitor frame. have. This colorful dynamic reverse phase is effective in blocking fine color and spatial patterns due to incomplete time integration. In a more general embodiment, the masking image may be generated as an image with dynamic inversion only in terms of luminance and the color arbitrarily changed.

In another embodiment of the present invention, the following private image separation method may be used at a special sequence position for smooth shutter opening and closing. For example, when one pixel image value of the private image is (R1, G1, B1) and the image P2 (R2, G2, B2) are displayed successively, instead of the image P3 having (R3, G3, B3) The images P4 having (R4, G4, B4) are displayed in succession. Here, the dynamic average pixel image values of the image of (R1, G1, B1) and the image of (R2, G2, B2) are the dynamic average pixels of the image of (R3, G3, B3) and (R4, G4, B4). It is equal to the video value.

When providing a private display to two or more authorized users simultaneously as a monitor, follow the formulation rules. A private video frame of user a is defined as P_a, a masking video frame of user a as M_a, a private video frame of user b as P_b and a masking video frame of user b as M_b. At this time, P_b and M_b also act as masking images to user a as M_a. Even when providing a private display to two or more authorized users at the same time, it is possible to combine and modify various combination rules as described above, using P_a, P_b, M_a, and M_b as image elements.

According to the present invention as described above it is possible to generate a more effective masking image in consideration of human visual perception characteristics. In addition, it is possible to generate a circular guided image and a disturbed image such as a dynamic reversed image in real time.

Claims (48)

In the device for outputting a private image using a public display, Means for generating a private video, Means for generating a masking image for masking the private image; Means for generating an image sequence pattern for the private image and the masking image; Means for outputting the private image and the masking image to the display according to the image sequence pattern; And the masking image generating means generates a dynamic reversed phase of the private image as a masking image according to the refresh rate of the display and the image sequence pattern. The method of claim 1, And a shutter opening and closing means having a shutter unit and a shutter control unit which opens and closes the shutter unit in accordance with a shutter opening and closing sequence pattern corresponding to the image sequence pattern. The method of claim 2, The shutter opening and closing means further comprises an ambient light blocking filter. The method of claim 1, The masking image generating means calculates an average image value Gad and then converts the average image value Gad. or - here, Is the dynamic inverse value, Is the average image value, Is the private video value, Is the maximum image value, Is the minimum image value, Denotes a gamma value of the private video output device, respectively, and denotes a dynamic inverse value for the private video. Private video output device, characterized in that for calculating. The method of claim 1, wherein the masking image generating means a) calculate an approximation of the mean image value, b) a predetermined pattern has an approximation of the average image value and the other portions are approximately minimum image values. Or maximum image value The image having a circular shape, the pattern having an approximation of the average image value, and the rest of the or The average image value is selected so that the sharpness of the pattern is the lowest while displaying by replacing the reversed phase having a half by c) - here, Is the dynamic inverse value, Is the average image value, Is the private video value, Denotes a gamma value of the closed image output device, respectively, to calculate an approximate dynamic inverse value, d) the pattern has the average image value and other portions or The image having a circular shape, and the pattern has the average image value, or The dynamic reversed phase value is selected by changing the reversed phase with half and displaying the pattern with the lowest sharpness. Private video output device, characterized in that. The method of claim 1, And the masking image generating means generates a masking image by further generating a disturbing image based on human visual perception characteristics and mixing the dynamic reversed phase. The method of claim 1, And said masking image generating means generates a dynamic reversed color image. In the device for outputting a private image using a public display, Means for generating a private video, Means for generating a masking image for masking the private image; Means for generating an image sequence pattern for the private image and the masking image; Means for outputting the private image and the masking image to the display according to the image sequence pattern; And the masking image generating means generates a disturbing image based on a human visual perception characteristic as a masking image. The method of claim 8, And a shutter opening and closing means having a shutter unit and a shutter control unit which opens and closes the shutter unit in accordance with a shutter opening and closing sequence pattern corresponding to the image sequence pattern. The method of claim 9, The shutter opening and closing means further comprises an ambient light blocking filter. The method of claim 8, And the masking image generating means further generates a masked image by further generating an induced image of the private image according to the refresh rate of the display and the image sequence pattern and mixing with the disturbed image. The method of claim 8, The masking image generating unit may include a disturbing image generating unit configured to generate a suitable disturbing image according to a disturbing image generating guideline based on a required visual perception characteristic. Private video output device characterized in that it comprises. The method of claim 12, The masking image generating means And a storage unit for setting and storing the index of the disturbed image based on the characteristic thereof. The disturbed image generating unit generates a disturbing image by searching for a suitable disturbing image from the storage unit using the index according to the disturbing image generating guideline. The method of claim 12, The disturbing image generating unit is configured to inspect the characteristics of the private image to establish the disturbance image generation instructions, characterized in that for establishing. The method of claim 12, The disturbing image generating unit generates a disturbing image covering a main spatiotemporal frequency of the closed image. The method of claim 12, The disturbing image generating unit generates a disturbing image that satisfies a predetermined frequency characteristic condition so as to include a frequency component sensitive to human visual perception at least a certain level. The method of claim 12, The disturbing image generating unit generates a disturbing image of which at least one of luminance contrast or color contrast of each image element constituting the disturbing image is greater than or equal to a predetermined value. The method of claim 17, The disturbing image generating unit may generate a disturbing image having any one of a luminance contrast or a color contrast of each image element constituting the disturbing image higher than the private image. The method of claim 12, The disturbing image generating unit generates a disturbing image by combining two or more images having sensitive frequency characteristics, contrast characteristics, and cognitive content characteristics. The method of claim 19, And the combination arranges an image having different characteristics for each frame region of the display. The method of claim 19, The combination is a private video output device, characterized in that for superposing the image having a different characteristic in one area of the display. The method of claim 12, The disturbing image generating unit may generate a disturbing image that is more sensitive to visual perception than a specific region of the display. The method of claim 22, The disturbing image generating unit generates a disturbing image in which a center portion is sensitive to visual perception compared to an upper portion or a lower portion of the display. The method of claim 12, The disturbing image generating unit generates a disturbing image including an image element having characteristics similar to the private image. In the device for outputting a private image using a public display, Means for receiving a private video from an external server or generating a private video by itself; Means for receiving a masking image from an external server, Means for generating an image sequence pattern for the private image and the masking image; Means for outputting the private image and the masking image to the display according to the image sequence pattern; Private video output device comprising a. The method of claim 25, And a shutter opening and closing means having a shutter unit and a shutter control unit which opens and closes the shutter unit in accordance with a shutter opening and closing sequence pattern corresponding to the image sequence pattern. The method of claim 26, The shutter opening and closing means further comprises an ambient light blocking filter. In the device for outputting a private image using a public display, Means for receiving a private video from outside; Means for generating a masking image for masking the private image; Means for generating an image sequence pattern for the private image and the masking image; Means for outputting the private image and the masking image to the display according to the image sequence pattern; Private video output device comprising a. The method of claim 28, And the video output means compresses a range of video values of the private video relative to the masked video. The method of claim 28, And the masking image generating means generates a dynamic reversed phase of the private image as a masking image according to the refresh rate of the display and the image sequence pattern. The method of claim 30, And said masking image generating means is made by an inverter for inverting said private image. The method of claim 28, And the masking image generating means generates a disturbing image based on a human visual perception characteristic as a masking image. The method of claim 28, And a shutter opening and closing means having a shutter unit and a shutter control unit which opens and closes the shutter unit in accordance with a shutter opening and closing sequence pattern corresponding to the image sequence pattern. The method of claim 33, wherein The shutter opening and closing means further comprises an ambient light blocking filter. In the device for outputting a private image using a public display, Means for receiving a private video from outside; Means for generating a masking image for masking the private image; Means for receiving an image sequence pattern for the private image and the masked image; Means for outputting the private image and the masking image to the display according to the image sequence pattern; Private video output device comprising a. 36. The method of claim 35 wherein And a shutter opening and closing means having a shutter unit and a shutter control unit which opens and closes the shutter unit in accordance with a shutter opening and closing sequence pattern corresponding to the image sequence pattern. The method of claim 36, The shutter opening and closing means further comprises an ambient light blocking filter. 36. The method of claim 35 wherein And the image output means compresses an image value of the private image relative to the masked image. 36. The method of claim 35 wherein And the masking image generating means generates a dynamic reversed phase of the private image as a masking image according to the refresh rate of the display and the image sequence pattern. The method of claim 39, And said masking image generating means is made by an inverter for inverting said private image. 36. The method of claim 35 wherein And the masking image generating means generates a disturbing image based on a human visual perception characteristic as a masking image. In the device for outputting a private image using a public display, Means for generating a private video, Means for generating a masking image for masking the private image; Means for generating an image sequence pattern for the private image and the masking image; Means for outputting the private image and the masking image to the display according to the image sequence pattern; The masking image generating means A color table storage unit including a color table for generating a masking image, and an image converter for generating a masking image with reference to the masking image generation color table Private video output device characterized in that it comprises. 43. The method of claim 42 And a shutter opening and closing means having a shutter unit and a shutter control unit which opens and closes the shutter unit in accordance with a shutter opening and closing sequence pattern corresponding to the image sequence pattern. The method of claim 43, The shutter opening and closing means further comprises an ambient light blocking filter. The method of claim 42, The color table storage unit includes a color table for generating a masking image for converting the private image into a masking image, and the image converting unit converts the private image with reference to the masking image generation color table to generate a masking image. Private video output device characterized in that. The method of claim 42, The masking image generating means A disturbance image generator for generating a disturbance image; An image converter configured to generate a masked image from the disturbed image by converting the disturbed image by referring to the color table for generating the masked image Private video output device characterized in that it comprises. The method of claim 42, A color table storage unit having a color table for generating a private image; An image converter configured to generate a private image by referring to the color table for generating the private image; Private video output device further comprises. The method of claim 47, And setting the image value ranges of the masking image generation color table and the private image generation color table differently to compress the range of the image value of the private image relative to the masking image.