GB2198842A - Movement sensing infra-red system - Google Patents

Abstract

A detection system sensitive to the movement of an infra-red source comprises an infra-red detector 1 and a grid 2 of alternately opaque 3 and clear 4 linear zones. The grid is placed to intercept infra-red radiation passing from a source of such radiation to the detector. A positive cylindrical lens 13 is placed between the source and the detector to concentrate radiation in the direction parallel to the grid zones to increase the radiation incident on the detector while preserving the light and dark striped shadow pattern due to the grid at the detector. Movement of the source produces movement of the pattern across the detector to give rise to an alternating output signal. <IMAGE>

Description

DESCRIPTION MOVEMENT SENSING INFRA-RED SYSTEM This invention relates to infra-red systems and, in particular, to an infra-red radiation detection system sensitive to the movement of an infra-red source comprising an infra-red detector and a grid of alternately opaque and clear linear zones, the grid being placed to intercept infra-red radiation passing from the source to the detector.

Such a movement sensing system may be used, for example, in an automatic light switching system or in an intruder alarm in which the movement of a person in a direction transverse to the grid causes the shadow of the grid cast by the radiating parts of the person to move across the detector. This gives rise to an alternating output signal from the detector as the bright and dark bands of the shadow pass in succession over the detector, the signal being used, for example, to switch lights on or off or to raise an alarm. Such a system is described in British Patent Specification 1,551,541. Therein a cylindrical mirror is used to focus radiation in a direction transverse to the grid zones, thereby expanding the field of view of the system.

It is an object of the invention to increase the radiation incident on the detector for a given source and detector.

The invention provides an infra-red radiation detection system as described in the opening paragraph characterised in that a positive cylindrical lens is placed between the source and the detector so that source radiation is concentrated upon the elements in the direction transverse to the cylinder axis of the lens. In the case of a cylindrical lens, the axis of symmetry of the lens lies in the lens surface and is parallel to the cylinder the cylinder generating axis of the cylindrical lens surface.

The action of the cylindrical lens is to focus radiation in the direction parallel to the grid zones but not in the transverse direction. The height of the lens in the grid zone direction can now be considerably larger than the detector height, a greater total amount of radiation being collected and concentrated on the detector than with the grid alone. Desirably, the pitch of the zones is constant across the grid. Then, a radiant object passing across the field of view at a constant velocity will give rise to a constant frequency alternating signal from the detector, which frequency can be measured to give information about the object movement.

The system may be characterised in that the grid and lens are adjacent and in that the separation between the lens and the detector is substantially equal to the focal length of the lens. The grid may then be provided on the surface of the lens, giving a simple structure to assemble. In practice, the separation between lens and detector may be slightly greater than the lens focal length. Then objects at a predetermined distance from the lens, large compared to the focal length, are sharply focused on the detector. The predetermined distance would, for example, be chosen to be the most probable distance of a person in a particular installation.

Desirably, the cylindrical lens is a thin fresnel lens for high transmission and for weight reduction. The system may then be characterised in that the fresnel lens is a plastics lens, and in that the grid and lens are adjacent and curved into an arc, the axis of which arc is parallel to the grid zones. The system may then be further characterised in that the arc axis is located at the detector, and in that the radius of curvature of the grid and lens arc is substantially equal to the focal length of the lens. The optical condition of lens and detector is then constant for the whole field of view covered by the arc of the curved lens. Alternatively, the grid and lens may be flat and the separation between lens and detector reduced somewhat below the lens focal length to give a less obtrusive, flat, shallow installation.Performance will still be optimum in those directions in which the lens to detector spacing is correct for focus, though less than optimum in other directions. In a further alternative an arc lens may be used with a flat grid, which could be placed outside the lens to mask its appearance, retaining the wide focused field of an arc lens.

To increase the detector signal the system may be characterised in that the detector comprises a pair of infra-red detection elements connected to provide an output signal indicating the difference in infra-red radiation incident upon the elements, in that the linear zones are transverse to an element separation line joining the centres of the elements, in that the cylinder axis of the lens is parallel to the element separation lines, and in that there is a gap between the elements along the element separation line. A bright band of the shadow then clears one element completely before falling on the other element, ensuring a maximum change in detector signal from the oppositely connected pair of elements.Then, desirably, the system is characterised in that the pitch of the grid is not less than one and a half times the element pitch and not greater than two and a half times the element pitch. Then, a bright band moves off one element at about the same time as it moves onto the other element. The change in detector output is thereby increased.

An embodiment of the invention will now be described, by way of example, with reference to the accompanying drawings in which, Figure 1 shows a schematic perspective view of an infra-red detection system in accordance with the invention.

Figure 2 is a side sectional view of Figure 1 with a representative object, Figure 3 illustrates the effect of variation of source width and/or range, Figure 4 shows the waveform obtained in an optimum case, and Figures 5 and 6 show typical installations of the system.

Referring to Figure 1, an infra-red radiation detection system is shown comprising an infra-red detector 1 and a grid 2 of regularly spaced alternately opaque 3 and clear 4 linear zones placed to intercept infra-red radiation passing from a relatively distant source 5 (not shown) to the detector. In this embodiment, the detector 1 comprises a pair 6, 7 of infra-red detection elements, for example pyroelectric detector elements, connected in opposition to provide an output signal at terminal 8 indicating the difference in infra-red radiation falling on the elements. Such a detector pair and circuit arrangements therefore are described in G.B. Patent Specification 2,O46,431B CPHB 32651). However, a single element detector may be used.

The linear zones are transverse to an element separation line 9 joining the centres of the elements so that the shadow of the grid cast by the source 5 moves in the direction of line 9 as the source 5 moves in the direction 10. As the source moves, alternate bright and dark bands of the shadow pass in succession over the detector. The pitch 11 of the grid can be chosen in relation to the pitch 12 of the elements, as will be described later, so that an alternating output signal is obtained.

A positive cylindrical Fresnel lens 13 is placed between the source 5 and the detector 1, the axis of symmetry 14, parallel to the cylinder generating axis of the lens cylindrical elements 18, being parallel to the element separation line 9.

The Fresnel lens and the adjacent grid 2 are formed into coaxial arc shapes, their arc axis 15 being parallel to the grid zones and located between the detector elements 6 and 7. The horizontal field of view of the system is defined by the angular subtense of the grid and lens at the detector, twice the angle 16 shown in Figure 1. Thus, from all directions in this field of view, the lens and grid are locally substantially normal to the line of view to the detector. If the source 5 moves with constant angular velocity about axis 15 , a constant frequency alternating output signal will be obtained. However, the arc axis may be located behind the detector to give a quasi-flat system. In the limit a flat lens and grid may give a conveniently shallow equipment with loss of performance only in those directions in which the lens/detector spacing does not give accurate focusing. A flat external grid could be used with an arc lens.

Referring to Figure 2, the radiation concentrating effect of the lens in the vertical plane is illustrated. A vertical section is shown through lens, detector and source. It is assumed that the source 5 is extended in the vertical direction, as would be the case with a person for example. Only that vertical extent h5 of the source is shown which is imaged by the lens on the detector element height hd, radiation from source parts above and below this vertical extent being imaged off the element and not contributing to the output signal. Also, in Figure 2 the distance r of the source from the lens is shown greatly reduced in comparison to rO, the lens to detector distance.In practice r would be of the order of one hundred times rO. The angle e is therefore thesemi-angular height of the source height h5 as seen from the installation.

In the absence of the fresnel lens, the apparent detector element height hd' at the plane of the lens and grid is given by: hd = hd + 2 tans. rO (1) The apparent detector element height in the presence of the fresnel lens is given by hg, the grid height, the lens collecting radiation over this increased height. Thus the optical gain G is given by: G = hg t2) hd It is assumed that all the radiation falling within hg is incident on the detector.

In a practical example, the effective source height h5 might be 1 metre seen at a range r0 of 5 metres, whence e = degrees. A practical detector might have hd = 2 millimetres and the lens to detector spacing rO might be 13 millimetres.

From equation 1, hd' is then 5 millimetres. A practical fresnel lens might have a vertical height, hg, of 14 millimetres, having an F number of 0.9. From equation 2 the optical gain is then 14/5 = 2.8. Thus a substantial gain in signal can be achievedby the use of a cylindrical lens. In practice, the lens/detector spacing may be reduced for compactness at the expense of some performance. It should be noted that there is no focusing action in the horizontal plane so that the width of the grid shadows cast by the source on the detector are unchanged, only their height being reduced.

The profile of the edge of the shadow bands cast by the source is determined by the angular width of the source as is shown schematically in Figure 3. With a wide source as shown in (b) the "umbra" 20 behind each opaque zone of the grid narrows sharply and the "penumbra" 21 widen rapidly with the distance away from the grid. The the shadow edges become blurred (not shown in Figure 3) and, eventually, the contrast between bright and dark bands would be degraded. In Figure 3(a) the source is more distant or of smaller width. The "umbra" 22 and "penumbra" 23 now taper more gradually and in the plane 24 of the detector the "umbra" and "penumbra" are more nearly equal in width than they are in Figure 3(b). But it should be noted that the pitch P of the shadow pattern in both (a) and (b) is the same, being determined by the pitch of the grid 2.The ratio of the widths of the opaque and transparent zones of the grid can be chosen in relation to the element widths and separation to provide a good contrast signal over a reasonable range of source angular widths or over a spread in range of a given source width.

In Figure 4, which shows an optimum choice of system parameters, the detector elements 6 and 7 are shown in front view, each element being of width W and separated by a gap of width W, the pitch of the elements being 2W. The bright and dark shadow bands are shown as idealised intensity graphs. The bright band is shown as the positive going part of the idealised rectangular bands I in Figure 4, the dark band being the negative going part. In the graphs the shadows are assumed to be sharp edged. It should be noted that the ratio of the widths of the bright and dark bands at the detector will change with source distance due to the finite source width as explained above. Also the contrast between light and dark bands at all source distances is degraded to some extent by radiation spilling over from bright bands into adjacent dark bands.

Before discussing the waveforms V that are obtained, certain properties of the detectors used should be noted. The pyroelectric detector used produces a signal voltage output which is proportional to the temperature change in the detector material produced by a change in radiation absorbed by the detector. Typically, changes in radiation at a frequency of about 0.5 Hertz produce optimum output. At frequencies as low as 0.1 Hz signals can be produced by ambient temperature changes or by air currents. It is therefore usual to provide the detector with an a.c. amplifier whose gain falls off rapidly below 0.5Hz, so that signals due to these causes are strongly attenuated.At frequencies materially higher than 0.5 to 1.0Hz, -the detector response falls off roughly inversely proportional to frequency, since the time period of the radiant energy change is much shorter than the thermal time constant of the detector.

In a typical installation, a person might move at 0.8 metre per second, transverse to the line of sight to the apparatus, at a distance of 5 metres. With a lens of focal length 12mm, the demagnification to the detector is about 400 times. Thus the speed of the image across the detector is 2.Omm per second. For an image movement of the order of 4mm, an a.c. signal of the order of 0.5Hz will be produced, assuming that the demagnified source width is of the order of a detector element width. These figures are merely given by way of illustrating the order of magnitude of the parameters involved in a typical installation.

In practice, variations in source velocity and range will produce a range of signal frequencies either side of the optimum frequency of 0.5Hz. Thus the signals shown in Figure 4 are considerably smoothed, due both to "penumbra" and to a.c.

frequency response effects.

Referring to Figure 4 the bright and dark bands are equal in width and each are equal to 2W, that is, twice the element width. The grid pitch is 4W. At t = 0 the bright band falls on element 7 and the gap, whereas element 6 is covered by the dark band. Element 7 is assumed to give rise two negative going signals and element 6 to positive signals on being heated. In this initial position the source is assumed to be stationary so that, due to the a.c. properties mentioned above, the output voltage is zero (a). It is then assumed that the source begins to move at velocity V, moving by a distance W in the succession of graphs (a) to (e) inclusive. At (b) element 6 is illuminated and element 7 is occluded. Thus both elements suffer the full change in illumination but in opposite directions.Being connected in opposition their signals add to give the sharp positive rise of signal shown in graph (b). At (c) the further movement of W does not produce any change in illumination conditions and a.c. effects dominate to produce the drop shown in graph (c). A further movement of W now reverses the illumination condition of both elements simultaneously producing the sharp negative signal of graph (d). A fourth movement by W does not produce any change in illumination conditions and the a.c.

coupling produces the return to near zero voltage shown in graph (e).

If the source were to continue in motion, reversals of illumination on the two elements would alternate regularly with periods of no change in illumination and an approximately sinusoidal steady state signal would be obtained having a periodic time T given by 4W T= V If the source were to stop shortly after the position shown at (e) a small positive overshoot would occur due to the small change in illumination conditions then occuring.

In Figure 4 it will be noted that there is a range of band movement W over which the signal changes due to a.c. effects only. Thus there is latitude for the edges of the bands to be blurred by the penumbra with small loss of signal amplitude.

Figure 4 illustrates the effect of source size and/or range variation in the dotted I and V graphs.

If the grid pitch had been chosen equal to the element pitch, both elements would be illuminated simultaneously and, after a movement W, would both be occluded. Thus the signals of the elements would cancel out and no signal would be obtained for continuous source motion. If the grid pitch had been chosen equal to 3 or more times the element pitch, simultaneous changes in illumination conditions on the two elements would not occur and there would also be larger source movements in which no illumination changes would occur on either element. Much smaller and much lower frequency signals would be obtained. Thus grid pitches from 1.5 times to 2.5 times the element pitch will give useful signals.

Figures 5 and 6 show two possible installations of the system. In Figure 5 the detection system 30 is installed over a door opening 31. The equipment is turned on its side so that the projected grid zones 32 from a succession of parallel line regions which would be crossed by a person entering or leaving the room. In Figure 6, the equipment 40 is installed in the orientation of Figure 1 but angled downwardly so that the projected grid zones 42 from a succession of line regions diverging radially from the door opening 41. The output signals in either case may be used to raise an intruder alarm or switch on lights in the room or both.

Claims (12)

CLAIM(S)

1. An infra-red radiation detection system comprising an infra-red detector and a grid of alternately opaque and clear linear zones, the grid being placed to intercept infra-red radiation passing from a source of such radiation to the detector, characterised in that, a positive cylindrical lens is placed between the source and the detector so that source radiation is concentrated upon the elements in the direction transverse to the cylinder axis of the lens.

2. An infra-red radiation detection system as claimed in Claim 1, characterised in that the pitch of the zones is constant across the grid.

3. An infra-red detection system as claimed in any one of the preceding claims, characterised in that the grid and lens are adjacent and in that the separation between the lens and the detector is substantially equal to the focal length of the lens.

4. An infra-red detection system as claimed in any one of the preceding claims characterised in that the cylindrical lens is a fresnel lens.

5. An infra-red detection system as claimed in Claim 4, characterised in that the grid and lens are flat.

6. An infra-red detection system as claimed in Claim 4, characterised in that the fresnel lens is a plastics lens, and in that the grid and lens are adjacent and curved into an arc, the axis of which arc is parallel to the grid zones.

7. An infra-red detection system as claimed in Claim 6, characterised in that the arc axis is located at the detector, and in that the radius of curvature of the grid and lens arc is substantially equal to the focal length of the lens.

8. An infra-red detection system as claimed in any one of the preceding claims, characterised in that the detector comprises a pair of infra-red detection elements connected to provide an output signal indicating the difference in infra-red radiation incident upon the elements, in that the linear zones are transverse to an element separation line joining the centres of the elementswin that the cylinder axis of the lens is parallel to the element separation lines, and in that there is a gap between the elements along the element separation line.

9. An infra-red detection system as claimed in Claim 8, characterised in that the pitch of the grid is not less than one and a half times the element pitch and not greater than two and a half times the element pitch.

10. An infra-red detection system as claimed in Claim 9, characterised in that the element widths are both equal to a gap between them.

11. An infra-red detection system as claimed in Claim 10, characterised in that the opaque and clear zones are equal in width.

12. An infra-red detection system substantially as described with reference to Figures 1 and 2 of the accompanying drawings.