A DISPLAY SYSTEM HAVING MOVABLE IMAGE FRAMES THIS INVENTION relates to a display system having frames of images which are movable, and, in particular but not limited thereto, the frames are movable according to the speed to a moving vehicle so that the images appear to be relatively still to viewers in the vehicle.

BACKGROUND OF THE INVENTION

Prior art display systems known to the applicant require viewers to be located in front of their screens and be relatively static to each other in order to clearly see the images on the screen. When the viewers move past a screen of a display system, especially if they are in a moving vehicle, they would either not be able to see the images on the screens or see only blurred images. Accordingly, the spaces in transport tunnels and other locations where tens of thousands of passengers and motorists travel therethrough have not been used for advertising and promotion purposes. Often the spaces between transit systems paths and walls and fixed features are minimal. The paths and the walls may be curved. Known prior art display systems can not be easily accommodated in the spaces. Current electronic display screens are rigid with a substantial thickness of driver circuitry, designed with linear facets. Vehicles such as trains travel through tunnels of 300M upwards. These tunnels vary the world over with some older tunnels having very uneven and dirty walls with services supplied along these surfaces. Modern tunnels are typically constructed with services on the ground and have relatively clean level sides with support deviations only. Thus, any display system must be able to reliably function in those tunnels and be easily accessible for maintenance.

OBIECT OF THE INVENTION

It is an object of this invention to provide a display system which alleviates or reduces to a certain level one or more of the above disadvantages.

SUMMARY OF THE INVENTION In one aspect therefore the present invention resides in a display system for displaying images. The system comprises a display screen formed of one or more

display panels, the or each said display panel having at least one display segment formed of spaced pixel units, and the pixel units being arranged to be controllably energisable pixels; detection means adapted to detect a position and/or speed of a moving object traversing across the display screen, and to periodically generate a detected position signal and/or speed signal; and a controller adapted to controllably energise said pixel units to form images and to shift at least one frame of the images across the screen based on the detected position signal and/or speed signal.

Preferably, the display screen is configured for mounting to a building structure or a wall surface in a tunnel or of a river bank. It is further preferred that the display screen or display panels can be shaped for conforming with a contour of the building structure or the wall surface. The contour may include a curvature.

The moving object may be a vehicle. The vehicle may be in the form of a train, a bus, a motor car or a boat.

In one form the detection means may include a tracking unit adapted to track position of the object and/or a speed measuring unit adapted to measure speed of the object.

In another form the detection means may include spaced light barriers and/or light sources arranged on the display screen or in a path along which the object travels and a time measuring unit for measuring time taken by the moving object to travel a distance between the barriers/light sources.

The controller may be adapted to calculate an image refresh rate based on the detected position signal and/or speed signal. The controller may also be adapted to determine the display panel(s) or the display segment(s) for image shift based on this signal(s). The system may have a bus member arranged along one edge of the display screen or along one edge of the or each said display panel for conveying power and/or control signals to said pixel units; and the controller arranged to supply said control signals to the bus member(s) for controlling luminance of the pixel units.

Preferably, said at least one display segment is formed of spaced elongate members and each elongate member has a conductive carrier supported therein and

spaced controllable pixel units mounted on said conductive carrier, the pixel units in the elongate members being arranged to form a matrix of pixels.

The spaced elongate members may be oriented vertically or horizontally. Preferably, the image shift is based on a distance determined according to the number of elongate members for image shifting.

The bus member may be arranged along one edge of the display screen, or along one or each of the opposed edges of the or each said display panel for conveying power and/or video data signals to said carriers; the bus member of the or each said display panel being in communication with the controller arranged to supply said power and/or video data signals to the bus member(s) of the or each panel/segment for controllably energising said pixel units thereof to thereby controlling luminance of the pixels.

In one form, each elongate member has a video control unit arranged to, in response to the video data signals, controllably energise said pixel units therein. In an alternative form, the elongate members are arranged in at least one group of linked elongate members, and one of the linked elongate member in the or each group has a video control unit arranged to, in response to the video data signals, controllably energise said pixel units in the linked elongate members.

Preferably, said controller has one or more panel control unit(s) configured to receive video signals from a video source and to transform the received video signals into modulated video data signals for conveying to the bus member(s). The video data may be superimposed on power for the pixel units.

The panel control unit(s) may have an initialisation arrangement for initialising the video control units in the elongate members to respond to the video data signals with corresponding addresses allocated to the elongate members. The initialisation arrangement may also be configured to determine screen or panel or frame pixel size.

The display system may have a suspension or support arrangement for suspending or supporting said elongate members in a vertically or horizontally spaced direction. The suspension or support arrangement may include said one or more of the bus member(s) and/or one or more suspension lines extending along the

or each said display panel. As the bus member(s)/suspension lines are relatively flexible, the display panels are foldable or rollable.

In preference, the display screen is formed of two or more of said display panels arranged side by side, and adjacent display segments of the side by side display panels may be arranged to be interconnectable. Thereby, the size of the display screen is selectively configurable. The elongate members may be formed with end flanges shaped so that one end of each elongate member in a display panel can be interconnected with an adjacent end of a corresponding elongate member in an adjacent display panel. Alternatively, a coupling member may be provided for interconnecting adjacent ends of side by side display panels.

The display screen may be based on that taught in the applicant's co-pending International Patent Application Number PCT/AU2006/001421 (Publication Number WO 2007/035992) filed on 28 September 2006. The complete disclosure of PCT/AU2006/001421 is hereby incorporated herein by reference. BRIEF DESCRIPTION OF THE INVENTION

In order that the present invention can be more readily understood and be put into practical effect reference will now be made to the accompanying drawings which illustrate one preferred embodiment of the invention and wherein:

Figure 1 is a partial schematic view ally of an embodiment of the large scale LED display system according to the present invention installed in a railway tunnel;

Figure 2 shows the position of the train illustrated in Figure 1 seconds later;

Figure 3 is a partial cut-away view illustrating views from inside the train of the moving advertisement images on the large scale LED display screen of the embodiment of the system shown in Figure 1 ; Figure 4 is a schematic diagram illustrating the shifts of the position of a virtual screen at different times;

Figure 5 is schematic time diagram showing frame and update shifting times of the virtual screen;

Figure 6 is a block diagram showing the use of train speed to control the virtual screen shift;

Figure 7 is a schematic diagram showing two consecutive frames in a content stream of an image on the display screen;

Figure 8 schematically illustrates the distribution of the horizontally spaced tubular display elements; Figure 9 illustrates frames of images on column (tubular element) one when the train is first detected by the system shown in Figure 1 ;

Figure 10 shows frames of images on columns (tubular element) one and two at time slice one;

Figure 1 1 shows frames of images on columns (tubular element) one to three at time slice two;

Figure 12 shows frames of images on columns (tubular element) one to four at time slice three;

Figure 13 frames of images on columns (tubular element) one to five at time slice four; Figure 14 frames of images on columns (tubular element) one to six at time slice five;

Figure 15 is a schematic diagram showing light sensors sending activation pulses for controlling frame shift;

Figures 16 is a schematic diagram showing activation of column one frames on first detection of a train;

Figure 1 7 is a schematic diagram showing activation of columns one and two at time slice two;

Figure 18 is a schematic diagram showing activation of columns one to four at time slice four; and Figure 19 is a block diagram showing main control circuit elements for the elongate display members of the display panel.

DETAILED DESCRIPTION OF THE INVENTION

Referring to the drawings and initially to Figure 1 , there is shown an embodiment of the large scale LED display system 10 according to the present invention. The display system 10 is installed in a train tunnel 100 of a mass transit transport network. The tunnel has walls 102 (one only shown) extending from a

ground surface 104. The walls 102 have curved sections 106. A train 20 travels along a railway track108 laid on the ground surface 104.

The display system 10 has a relatively long display screen12 supported on the tunnel wall 102. In this embodiment, the screen 12 extends substantially the full length of the tunnel wall 102.The display screen 12 receives power and control signals from bus members 14 and 1 6, and in this embodiment it is formed of multiple number of interconnected display panels 18 (see Figure 3) which are substantially as taught in the above-mentioned co-pending patent application. Each display panel has a number of vertical spaced elongate members 22 (80 in said co- pending patent application) which contain LED pixel units (50 in said co-pending patent application). The elongate members 22 are arranged so that the display panels 18 can bend for conforming to the curvature of a curved section 106.

As shown in Figure 2, the train 20 travels along a track 108 and in a split of a second moves past a number of the display panels 18. Passengers in the train 20 would not be able to clearly see static images on the screen 12 because the train is in motion.

The inventors have envisaged visual effect of a motion image displayed at close range to mass transport systems, such as trains. The display could preferably be placed on existing features, such as in transport tunnels, blanketing the outside wall of the viewer. The display, stretching the entire length of the tunnel or structure, would roll the image along with the mass transit system, such that the image sequence would appear to be played directly outside the transit system windows. The display screen taught in the co-pending patent application incorporates a low profile 'tube'/elongate member layout, where by low power, low profile circuitry is encapsulated in a poly carbonate tube. The need for large power supplies and heavy equipment is eliminated at the screen face, leaving a bare minimum of decoding circuitry. This provides a low profile, modular unit, which is not restricted to the planar faces that similar displays are. Relatively small tubes ensure that the display screen 12 can mould to curvatures in the course and wall without having to modify the structure in any way.

A full matrix arrangement of LEDs, although could be utilised, would not be optimal for this situation. Due to the motion of the train, and the magnitude of the display screen 12, a sparse horizontal, and dense vertical arrangement as shown in

Figure 3 is much more desirable. This can be achieved by forming a large horizontal spacing between high density tubes 18 along the face of the wall 102.

The system 10 is arranged tp provide a relatively small virtual screen in the display screen 12 which linearly moves with the viewers. The display or physical screen 12 can be configured to stretch the entire length of the tunnel 100 or a part thereof. At any one time, however, a viewer is only able to see a small section of the display. This section is defined hereinafter as the virtual screen. A smaller screen image within the physical display, which may be of the same height, but only a fraction of the length, the image can be shifted along the display 12 with the viewers as schematically shown in Figure 4. This figure shows that the virtual screen 24 has been linearly shifted to three different locations along the display screen 12. For clarity purpose, this figure shows a single virtual screen 24 on the screen 12. In practice, the screen 12 has a number of virtual screens 24.

To achieve the shifting of the virtual screen 24, the physical display must know where the viewer is. This embodiment of the system 10 uses detection means 26 (see Figure 15) having light sensors 28 associated with corresponding tubes 22 to detect position of the train 20 relative to the display panels 18 to control shifting of the virtual screen 24. The light sensors 28 form light barriers or curtains for measuring train position. Alternatives include utilising tracking and speed measurement devices already in place on the mass transit network, or a method similar to that used in laser guided computer mice, where reflections are recorded from the objects surface.

Once position and time information (velocity) has been gathered, it is then processed to calculate the "motion-refresh rate". The motion-refresh rate (given in Hz) is the number of times the display must update per second to shift the virtual screen 24 with the viewer. The image is shifted by a set distance along the display screen 1 2 each motion-refresh. However the time taken before each shift is variable, and based on the speed of the train 20.

The number of tubes 22 the images shift by each time is a function of the minimum refresh rate, resolution and the minimum speed of the traint in the tunnel 100: v _min x h n = ^■

*mm

Where: v _mιn = Minimum velocity of the transport in meters per second r _mιn = Minimum acceptable refresh rate of the display n = Number of tubes shifted per motion refresh h = Horizontal resolution in tubes per meter The horizontal resolution of the image is ultimately determined by the number of tubes 22 present per meter in the display screen 12, which can be translated to a distance, d, representing the spacing between the tubes 22:

-ϊ

The velocity, number of tubes 22 shifted, distance between tubes 22 and micro refresh rates are linked by the following formulae: v

Motion Refresh Rate (T _nOt10n ) =

Where: v = Velocity of the transport in meters per second

The motion refresh rate, the rate of update, is no longer dependent on the actual input refresh rate of the display system 10. This is because the update of the display 1 2 is linked to the movement of the viewers/train, to maintain tracking. The frame changes of the input stream would be preferred to fit into the refresh of the display 1 2. This suggests that frames will need to change slightly ahead or behind of the average change point. The conceptual time diagram shown in Figure 5 illustrates this.

As shown in Figure 5, frames 1 and 2 are displayed twice at two different positions, as the train 20 accelerates to a maximum speed, and frame 5 is displayed 3 times at 3 different positions. This diagram illustrates how the dependency only

on train speed for the motion refresh rate combines with the refresh of the input refresh rate. The position of the frame change is on average 50% behind the actual frame change, varying from only a fraction behind, to almost the next frame. The transition between frames 7 and 8 occurs neatly at the change of shift, while the change from 8-9 occurs slightly after the shift transition and hence the frame is not updated until close to the end of frame 9. The maximum shift is just under 1/60 ^th of a second from the original frame.

This asynchronous method of frame transition resolves the issue of achieving shift speeds which are not multiples of the frame rate, without slippage or distortion. Differences of calculated and actual position at any point in time will cause the image to 'jitter' or shift rapidly back and forth in the horizontal direction relative to the viewer. The prominence of this effect is depend both on the sampling rates of the train speed and the amount of shifting per micro refresh. Ultimately to minimise the misalignment and 'jitter' of the image, the minimum number of tubes 22 shifted, and minimum distance between tubes 22 can be used for any given velocity. This ultimately increases the micro-refresh speed of the display which is exponentially related to the cost of the constituent components, such that any small increase in the sample and refresh rates incurs a significant cost on to the display 12.

This can be balanced by utilising digital filtering tools, such as moving average and hysteresis type functions which help to buffer errors generated by hardware in the position calculation of the display 12. To minimise the distraction of the effect, it is possible to always move in one direction such that image appears to 'slide' one way, rather than 'jitter' back and forth.

Figure 6 shows a controller arrangement 30 for controlling images on the screen 1 2. The controller 30 has a train speed/position sensing unit 32 and a video signal reception unit 34. The two inputs, one from the video signal, the other from the velocity or position of the train, can be either processed from physical sensors or, extracted from already available information. The received video signal is stored in a buffer 36 and the speed/position information obtained by the sensing unit 32 is processed by a position calculation unit 38 with reference to a corresponding number of tubes 22. The video signal in buffer 36 and the position information

calculated in 38 are processed by an image processor 40 for producing tube control signals to a tube control unit 42. The control unit 42 distributes control signals to panel controllers 44 which control illumination of LED units 50 in display panels 18. The tubes 22 are seen running top to bottom of the display screen 12 when viewed from the front, as opposed to left to right. The screen 12 refreshes horizontally across the columns of tubes 22. To provide minimal latency and processing between the video signal and the display 12, the input image is rotated 90 degrees to compensate, allowing for the video to be transmitted using the vertical standard.

The train speed can be calculated in a number of ways, and this will depend greatly on the physical challenges of the specific display and the information already available. Information could be obtained from physical light interaction with the train body, through to communication with central information servers of the transport network. Light interaction could be performed using evenly spaced light beams passing perpendicular to the direction of travel, disrupted by the train 20 as it passes through. Or could be measured from reflections similar to how computer mice interact with the surface. Information from central servers would require a communication link between the display system 10 and the server. To give the illusion of a static display to a moving object, the information gained from the speed of the train 20, along with frame information from the video stream will be combined to place the image at the correct place in the tunnel 100. Figure 7 shows two pixelated images in a sequence, Frame 1 and 2, represent two consecutive frames in a content stream to be displayed for a moving object. The frame rate of the input stream is 60Hz, the tubes 22 are placed at 2 times the vertical resolution (say 1 50mm apart) and the train 20 to be tracked is moving at 40km/h. Since the train 20 is travelling at 1 1 m/s, and each tube 22 is placed at 50mm, the train 20 passes 222 tubes every second at any one point. As the refresh rate is 60Hz, each frame is 'passed' along to the next tubes 22 almost 4 times (3.7), before changing to the next frame on the succeeding pass. The first ten tubes are shown in Figure 8. Once the train 20 is detected at the start of the screen display 1 2

the controller 30 would process to obtain the train speed/position and display the image on the far right column of Frame 1 (Column F), on Tube 1 as shown in Figure 9.

A second moment in time later, at approximately one quarter of the refresh time T = 1/222s, (Since the transport will pass 222 LED columns per second at any point) the frame has shifted up one tube, making space for the second pixel column

(Column E) to make an appearance in physical LED Column 1 as shown in Figure 10.

The process continues to shift in column D of Frame 1 as shown in Figure 1 1 , and continuing, for the last time, to shift in Column C of Frame 1 a shown in Figure 12.

The next shift will be in time for Frame 2, since 4/222 is approximately 4/220 (220 = 4Shifts X 60Hz), hence all columns will then change to Frame 2 as shown in Figure 13. The process continues to shift in the remaining section of Frame 2 in the next time slice as shown in Figure 14: The process then continues to shift Frame 2 across the display 12. The micro refresh rate of the tubes 22 here is 222Hz to keep up with the train 20, This can be reduced by not updating the display at every possible instant.

Removing time slice 1 ,3, and 5/222 s from the above diagrams can result in the display shifting 2 tubes across, at half the speed. The visual effect of this causes some quality lost. However granted that visual perception of LEDs and similar displays are only around 60Hz, 1 1 1 Hz update speeds may not cause any discomfort at all.

In this display system 10, each tube 22 displays more than one column of any frame, and image may shift mid frame. To take advantage of this, all tubes 22 are sent the entire frame, and no tube will be individually addressed.

As shown in Figurei 5, master tubes 22 with references 1 through 10 are sent the entire frame, buffered to memory in the master tubes 22. As the train 20 passes light sensor 28 (of corresponding master tube referencel ) the light sensors generate a pulse transmitted to the Activation Pulse Input of the corresponding Master tube, as well as a pulse to the Clock Bus, received by all other Master tubes on the Shift Pulse Input.

A section of the frame, exactly the correct amount of pixels wide to match the number of tubes 22 the master controller is responsible for. I.e., if there are 3 slave tubes plus a master, a frame slice would be 4 pixels wide. Since each tube 22has been provided exactly the same data, the tube relies upon the timing of the train 20 to display the correct image section. An "Activation Pulse" is a pulse sent from the light sensor 28 directly connected to the master controller. It signifies the transport has passed the tube for the master controller to begin displaying the sequence of frame slices, starting the far right slice.

A pulse received on the Clock Bus from any other light sensor 28 along the tunnel 100. This indicates that the train 20 has moved another frame slice along and the image should update to match this, i.e. shift. For every shift pulse, the tubes 22 display the next frame slice available in memory, moving left across the image, until there are no frame slices left.

The Shift Pulse Input is connected to the Clock Bus, which in turn is connected to every single light sensor, this means that even before a tube has received an activation pulse, it will be receiving pulses from the Shift Input. These will be ignored, until an activation signal is received. When a new frame is available, it is sent asynchronously to the movement of the display screen 12, overwriting the previous image in memory. The frame slice number that the tube 22 was currently displaying is stored separately, such that on the next update, the incremented slice of the new frame is displayed.

This method of updating eliminates the need for addressing the columns, ensuring that the large tunnel length can be accommodated by the communications structure, while significantly reducing the protocol overhead. Figure 19 shows the main components of the controller 30 in association with the tubes 22.

In Figure 1 6, it is seen that the train 20 breaking the light barrier at Master tube referenced 1 , causing an activation pulse at tube referenced 1 . Master tube 1 displays the first three pixel columns of frame 1 . Every other tube receives a shift pulse, however those which have not received an activation pulse, remain dark. When the train 20 crosses the light barrier connected to master tube referenced 2 as shown in Figure 1 7, tube 2 activates and tube 1 receives a shift pulse,

which appears as if the image has shifted 3 tubes to the right with the transport. Note that the frame has not changed as yet, however during the time it takes for the transport to reach the light barrier for tube 4, the frame has been updated to frame 2 as shown in Figure 18. The new frame (Frame 2) transmitted between update 3 and 4 has replaced frame 1 , such that when the transport passes the light barrier at tube 4, it displays the first 3 columns of frame 2, not frame 1 . Similarly, Master tubes 1 , 2, and 3 use the next respective block of 3 from memory containing frame 2. In this manner, the update speeds across the display screen 12 are inherently independent of the input frame rate. This may not be the case for every display system 10, as some display systems may utilise information already available from the train system network, however will operate on a similar refreshing method.

Whilst the above has been given by way of illustrative example of the present invention many variations and modifications thereto will be apparent to those skilled in the art without departing from the broad ambit and scope of the invention as herein set forth in the following claims.