WO1989006782A1

WO1989006782A1 - Method and device for scanning an object

Info

Publication number: WO1989006782A1
Application number: PCT/SE1989/000006
Authority: WO
Inventors: Johan Forssblad; Per Andersson; Bo Lindberg
Original assignee: Emve I Åstorp Ab
Priority date: 1988-01-12
Filing date: 1989-01-12
Publication date: 1989-07-27
Also published as: SE462696B; NL8920012A; DK159190D0; GB2232765A; SE8800074D0; SE8800074L; DK159190A; GB2232765B; GB9015071D0; FI903480A0

Abstract

The present invention relates to a method and a device for contactless scanning of an object (1). Light is conducted by distributing means (11, 12, 13) to the object and neighbouring surfaces (2, 16). A detecting means (9) detects light which is reflected by the object and the surfaces. The reflectivity of the surfaces is excellent, and the detecting means receives reflected light of high intensity from the surfaces. Light hitting the edge portion of the object will be spread when reflected, which implies that the detecting means receives light of low intensity. Consequently, the intensity will be high for the surfaces but will be at its lowest at the contour of the object. Therefore, signals depending on the detected intensity can be processed in a computer. The computer calculates a reproduction of the object, which can be used for e.g. calculating the length, width, height or volume of the object.

Description

METHOD AND DEVICE FOR SCANNING AN OBJECT

The present invention relates to a method for contactless scanning of an object positioned adjacent or on a surface, said method comprising emitting light in the form of a light beam towards said object, scanning said light beam across said object, detecting the light after reflection by said object and said surface, determining the intensity of the detected light, thereby determining the contour of said object. Technical field of the invention When treating a great variety of objects in, for example, flow line, there are several important factors to be considered. It is frequently desirable to measure in a simple way the dimensions of an object and determine the size, weight and volume. In many cases, it is also important to determine the exact shape of the object to allow it to be sorted into a group of similarly shaped objects. If a plurality of objects are positioned on a surface, such as a conveyor belt, it can also be of great interest to determine the relative position of the objects on the surface. Objects that are to be scanned, measured and determined may differ greatly from each other. It is difficult to distinguish from a mixture of screws and nuts the separate objects for the purpose of sorting, possibly except in manual work. Background of the invention

To make it possible to keep the prices of vegetables at a moderate level and still obtain a satisfactory profit, it is important for growers and wholesalers to sort the vegetables according to shape and size so as to make each lot uniform. For example, tomatoes are sorted according to shape. Small and round tomatoes are sorted separately, while large and differently shaped tomatoes are sorted separately. When it comes to root-crops like potatoes, there are many different requirements. Restaurant kitchens desire small and round potatoes, while companies manufacturing potato crisps want long and thick potatoes, and the individual consumer wants above all lots of similarly and uniformly sized potatoes.

The possibility of separating, from a large number of objects, objects which somehow deviate from a certain, fixed standard is important. Green tomatoes deviating from the red ones may be involved, rusty screws among the clean screws or the wrong type of vegetables. Especially in connection with potatoes where potato-shaped stones may occur, it must be possible to simply identify the objects. For scanning, measuring and determining individual objects in a great variety of objects, all objects are in many cases positioned on a surface or a conveyor belt to be sorted. For quick and economic handling, it is convenient to register the position of the indi- vidual object on the surface, at the same time as the object is identified. One wants to register the position of each object in relation to a reference portion of the surface (e.g. the edge of a conveyor belt), and other neighbouring objects. There are various ways of sorting objects, and principally three basic ideas can be distinguished: manual, mechanical and contactless sorting.

The manual sorting simply requires operators standing adjacent the surface on which the objects are positioned, for sorting the objects. Besides being time-consuming and thus costly, this is an inaccurate method. Moreover, such moving line work means that it may be difficult to fulfil the requirements for a satisfactory working environment. The mechanical sorting can be effected in many ways. There are prior art techniques of sensing the weight of the object by means of levers, springs and the like. Here is, however, a certain risk that different kinds of objects which, however, have the same weight are sorted together. For potatoes, especially screen sorting is used. The potatoes are sorted by means of a screening device comprising superposed screens with screen meshes which in Sweden are square. The potatoes are shaken out on the screening device until they slip through a screen having a suitable mesh size. Finally, the potatoes reach a screen having such a small mesh size that they cannot slip through. If, for example, a potato can pass a 55x55 mm mesh but not a 35x35 mm mesh, this is a size 35/55 potato. There is a certain risk that elongate potatoes are mixed with small and round potatoes, since the largest cross section of the elongate potato has the same dimensions as the ball-shaped potato.

Such mechanical sorting is not quite accurate and may damage the objects and result in a mixture of objects of different sizes. The contactless methods can be optical, electrical etc. One method uses a video camera having a CCD picture tube for reproducing each object. Since no operator should sit in front of a video screen, picture analysing is used. Here, for example colour deviations can be detected. The problem of this method of sorting objects is that picture analysing is both difficult and costly and yields unreliable results.

A device according to US 4,025,796 uses a wide and thin beam of light which scans across a fixed object. The reflected light is detected, and the height of the object is determined. The problems are that moving objects cannot be measured and that only one measure of the object is obtained.

Swedish Patent No. 8403364-6 discloses a method and a device for contactless detecting of plants.

Light which has been reflected by plants and a surface is collected and separated in two luminous fluxes each having a narrow wave range. An electric circuit processes the light signals which have been detected by means of the device, and produces an output signal which indicates the presence/non-presence of a plant on the surface. Such detecting is, however, rather primitive and cannot directly furnish more detailed information on the position or the character etc of the plant.

The main object of the present invention is to provide a method and a device for contactless scanning of an object, thereby solving the above-mentioned problems.

A first special object is to determine the shape of the object such as the shape is defined by the contour of the object for at least two different directions.

A second object is to determine the weight of the object.

A third object is to determine the volume of the object.

A fourth object is to determine the centre of the object, which in certain ideal geometric shapes of the object can coincide with the centre of mass of the object. A fifth object is to determine the position of the object on a surface in relation to a given reference, and optionally in relation to other neighbouring objects.

A sixth object is to identify objects, and parti¬ cularly to distinguish different objects of similar shape.

A seventh object is to simultaneously scan two objects positioned on the same surface or on separate surfaces.

According to the invention, the above objects are achieved by a method and a device which are defined in the characterising clause of claim 1 and 7, respect¬ ively. Preferred embodiments are stated in the dependent claims .

The invention will now be described in more detail with reference to the accompanying drawings.

Fig. 1a is a plan view of an object on a surface, Fig. 1b is a side view of the same object as seen from a line A-A, and Fig. 1c is a side view of the same object as seen from a line B-B;

Fig. 2 is a schematic perspective view of a pre¬ ferred embodiment of the device according to the invention;

Fig. 3 shows the beam path for the device in Fig. 2;

Fig. 4 shows an object lit by parallel light, and a diagramme of the intensity of the light reflected by the object;

Fig. 5 is a perspective view obliquely from above, showing ^'how an object adjacent a surface is scanned ^• by light from two directions;

Fig. 6a is a diagramme of the analogous signal which is produced in detecting of the reflected radiation, and Fig. 6b is a diagramme of its corresponding digital pulse train; and

Fig. 7 is a block diagramme of the signal processing according to the invention. Reference is first made to Figs 1a-1c. Fig. 1a shows an object 1 positioned on a surface 2. In this example, the object resembles a stone. Fig. 1a is -a plan view of the object, while Figs 1b and 1c are side views from the lines A-A and B-B, respectively. In optical scanning of the object, the dimensions of the object are to be determined. The largest length of the object is indicated by the designation 3a, while the largest width of the object, which is perpen¬ dicular to its largest length, is the sum of the arrows designated 3b and 3c.

When studying the object as projected on the line A-A, one finds the figure as shown in Fig. 1b, and it will be appreciated that the maximum length 6 shown in this Figure is not the true length of the object. An analogous discussion applies to the projection of the object on the line B-B. The Figure also shows the centre of mass 7 of the object, which in ideal cases coincides with the geometric centre of the object.

Then reference is made to Fig. 2. In a preferred embodiment, several objects 1 are positioned on a surface 2 which in this case may be one of two parallel, endless conveyor belts. The belts each feed the objects through a gate in a scanning device 30 designed according to the invention. A light-emitting means 8 comprises at least one light source which emits a thin beam, preferably of a diameter of about one mm. A rotating polygonal mirror 11 having light-reflecting facettes reflects the emitted light. The light is directed by special lenses 14 and 15, so called fθ lenses. The lense 15 constitutes part of the side wall of the gate, and the lense 14 part of the roof of the gate above the surface 2. A second surface 16 constitutes the second wall of the gate. The light reflected by the belt 2, the surface 16 and possibly the object 1 is conducted back to a detector 9 which comprises at least one photodetector, for example a photodiode. The output signal produced by the detector 9 is conducted to an electronic circuit and computer unit 17.

Reference is now made to Fig. 3. The light-emitting means 8 preferably comprises a helium-neon laser. Also other light sources can be used, e.g. light bulbs, flash discharge tubes or fluorescent lamps. The main thing is that a substantially parallel light beam is produced. Since a laser in general and the helium-neon laser in particular has an extremely small light beam divergence, laser light is suited for the present purpose. The laser light beam is emitted from the laser towards the rotating polygonal mirror 11 via a mirror 10. The mirror 10 either has a bored hole for letting through the laser beam, or is semitransparent.

In the preferred embodiment, the rotating polygonal mirror 11 has six facette surfaces capable of mirroring the light. When the rotating polygonal mirror 11 is rotated, a facette is moved into the beam of light which will then successively illuminate the facette surface of the polygonal mirror. Since the angle of incidence of the light beam towards the facette now changes successively, the light is successively spread in different directions, depending on which part of the facette is hit by the laser light. The light re¬ flected by the rotating polygonal mirror 11 is emitted towards two distribution mirrors 12 which are positioned above the two surfaces 2. The light is reflected by each distribution mirror 12 so that part of the light is substantially emitted straight down to the surface 2, and part of the light is emitted each towards a side mirror 13.

The light reflected by the side mirror 13 is emitted towards the surface 16 which extends substan¬ tially perpendicular to the surface 2. To ensure a parallel light beam, the side lense 15 is a so-called fθ type lense. As the rotating polygonal mirror 11 is rotated, a thin light beam will thus be scanned across the surface 16 and the object 1 , if positioned in the path of the beam on the surface 2.

The light emitted straight down to the surface 2 by the distribution mirror 12 is also corrected by means of the fθ lense 14 positioned above the surface 2. The object will be scanned also from this direction. The beam path 1 is screened in the right-hand part of Fig. 3. It is apparent that the object 1 is illuminated by a thin light curtain from the side and a thin light curtain from above. The object thus leaves a shadow on the surface 16 and the surface 2, respectively.

Reference is now made to Fig. 4 which is a cross- sectional view of an object 1 on a surface 2 which is illuminated by parallel incident light beams 190-194. Fig. 4 shows the situation when the thin light curtain in Fig. 3 is used, and explains how it is possible to detect the contour of -the object. The Figure also shows a curve of the intensity of the reflected light which can be perceived by a person in a position at right angles above the surface 2, i.e. at the source of the light beams 190-194.

When the planar surface 2 reflects the light 190, the intensity of a reflected light beam 200 is maximal, and this is best seen to the left in the Figure. The incident light beam 190 is in fact reflected in accordance with optical reflection principles sub¬ stantially perpendicular back and causes an extremely high intensity called the white level 22. In the preferred embodiment shown in Fig. 3, the device is however slightly inclined relative to the vertical plane. As a result, too strong a direct reflection from the surface 2 is avoided. The white level 22 is maintained as long as light is reflected by the surface 2.

When, the light beam 191 hits an edge of an object ~ _r the reflected beam 201 is emitted in a quite different direction. The observer will perceive almost no light at all. The intensity has been reduced to a so-called black level 21.

As the incident light beam is moved across the object 1 , more or less light will be reflected in such a direction that part of the light may be perceived by the observer. The light beam 192 which hits the object substantially perpendicularly will cause a strong reflected light beam 202 which also depends on the reflectivity properties of the object. A white egg could thus cause a strong reflected intensity, whereas a potato soiled with earth only provides a slight increase of the intensity as compared with the black level 21. The intensity curve provided under reflection by the object 1 is designated 23. The important lesson which is illustrated in Fig. 4 is that precisely at the outer edges of the object, the perceived intensity is minimal and contrasts strongly with the white level. This "edge effect" can be accen- tuated if the surface 2 is made of a strongly reflecting material, such as a metal strip or a white plastic mat.

Fig. 4 also shows by a dashed line a reference level 40 below which a safe minimum level, i.e. a black level, can be determined. Reference is again made to Fig. 3. The part of the reflected light which, as said above, can be perceived by the observer in a position at right angles above the surface 2, is collected by means of the lenses 14, 15 and reflected by the distribution mirror 12 to the rotating polygonal mirror 11 and further to the mirror 10 where the reflected light is deflected towards a detector 9.

In the preferred embodiment, a light beam which in relation to the sizes of the objects is narrow, is emitted from the light source 8 and scanned, by means of the rotating polygonal mirror 11, across the object, as shown to the right in the drawing by means of a screened beam path. The light beam preferably is a laser beam having a diameter of about one mm and a small divergence. The reflected light will go almost exactly the same way back, i.e. the same way as the incident light, except in the mirror 10 where it is deflected towards the detector 9.

In the preferred embodiment with six facettes on the rotating polygonal mirror 11 , six scannings of each object are effected under one revolution of the rotating polygonal mirror. Because of the rotation, each facette will turn the light beam from a position H at the extreme right in Fig. 3 to a position V at the extreme left in Fig. 3. Then the light beam passes to the left part of Fig. 3 and illuminates the other object. Reference is now made to Fig. 5 which illustrates how the object 1 is scanned by the incident light beam. The surface 2 is a web on which the object 1 is positioned. The web moves forward, and the incident light beam moves either transversely across the web and is directed perpendicularly to the plane of the web as shown at A, or in parallel with the plane of the web and is directed perpendicularly to the surface 16 as shown at B. Depending on the speed of the web and, respectively, the scanning of the incident light beam, the object 1 thus is divided into "slices". The slices shown are excessively thick. In actual practice, the thickness is equal to the diameter of the light beam. _- At A, the object is divided from above into a plurality of neighbouring slices, an observer thus seeing from above the thickness and length of each slice, i.e. seeing the contour but not the height of the slice. At B, the object 2 is divided from the side into slices. An observer can from the side see the height contour of the object, but cannot form a direct opinion of the width of the object.

In the preferred embodiment of the present invention, the object 2 is divided in this manner, the dividing being effected, largely alternately, at the same time from above according to A and from the side according to B.

The reflected light beam which is the result of the scanning by means of the incident light beam, is, according to Fig. 3, observed from a point at the light source. The intensity curve shown in Fig. 4 thus corre¬ sponds to a "slice" of the object and the web as taken from one direction. Several such intensity curves corre¬ sponding to several scannings observed after each other, together produce a reproduction of the object. Since the same object is scanned successively from two different directions, an observer receives an approximate repro- duction in three dimensions of the surface 2 and the object 1. It will be appreciated, however, that the reproduction is not a true three-dimensional reproduction, Then reference is made to Figs 6a and 6b. When the preferred embodiment of the invention is used, the detector 9 will, for each facette on the rotating polygonal mirror 11, produce a curve equalling the one in Fig. 6a. The digitised equivalent is shown in Fig. 6b. The reference numeral 24 designates the intensity curve obtained when the light beam is scanned in the vertical direction from above downwards across the right-hand object 1 in Fig. 3. The reference numeral

25 indicates the intensity curve obtained when the light beam is scanned across the object 1 from the right to the left in Fig. 3. The reference numeral

26 indicates the intensity curve obtained when the light beam is scanned across the left-hand object

1 from the right to the left. The reference numeral 27 indicates the intensity curve obtained when the light beam is scanned in the vertical direction from below upwards across the left-hand object 1 in Fig. 3. The interval 24 begins to the left with the high white level 22 which is obtained the first time the light beam hits the surface 16. When the light beam has moved a certain distance which corresponds to a period of time 28, the object 1 is hit, and the intensity sinks drastically to the black level 21. In the embodiment shown, the intensity increases as the light beam scans across the object, before the intensity again decreases to the black level. The distance between the points where the black level is obtained, represents the projected height of the object. After the object has been scanned, the surface 16 is again hit, if the object does not rest entirely on the surface 2 and thus conceals the surface 16.

Between the interval 24 and the subsequent interval

This digital signal is fed to two logically operated shift registers 106 and 107. A pattern detector 108 and 109, respectively, is connected to each shift register 106 and 107, respectively, and together these components constitute two parallel sequential networks 106, 108 and 107, 109. To one sequential network 106, 108, the signal from the comparator is fed, while to the other sequential network 107, 109, the inverse signal is fed. The purpose of the pattern detectors is to identify the patterns of certain predetermined pulse trains. The pattern detectors produce an output signal only when these patterns are identified in the signal. Figs 6a and 6b show at 25 and 26 how two different pulse trains can each represent an object. At 25, the front edge of the object causes a pulse trailing edge 50, and the rear edge of the object causes a pulse leading edge 51. At 26, the front edge of the object causes a pulse trailing edge 60, and the rear edge of the object causes a pulse leading edge 63. Moreover, the excellent reflectivity of the object gives an intensity which locally exceeds the reference level 40. As a result, a further pulse with a pulse leading edge 61 and a pulse trailing edge 62 is obtained. When one of the pattern detectors produces an output signal, this is fed to a buffer memory 112. To the buffer memory, a time value from a time base 110 is also fed, said time base being trigged by a synchronising start signal from a photodiode 111. The photodiode 111 is affected by light once for each facette of the rotating polygonal mirror. This means that the time base is zeroised for each facette, i.e. for each measuring cycle which comprises scanning of two objects from two different directions.

The time value corresponds to the specific angle which the rotating polygonal mirror takes for a certain front edge or rear edge of the object. These numbers are fed in line through the buffer memory. The computer 113 collects the numbers from the buffer memory which are required for determination of the objects. The computer 113 is connected to an output unit 114 which can be a character display or the like. The computer is further connected to a sorting means 115 to which a control signal is fed from the computer for performing sorting functions.

By means of the numbers comprised by the buffer memory, the computer decides which numbers belong to which object. In the preferred embodiment in which two objects are scanned simultaneously, the data of the objects are divided into two separate registers. Below, the operation of one register will be described. Both registers will be operated in the same manner. The next step for the computer is to combine the produced associated numbers into distances which represent the width and .height of the object, and then form "slices", the thickness of the slice being determined by the distance between each scanning, as shown in Fig. 5. The distance will, of course, depend on the speed of the belt and the speed of rotation of the rotating polygonal mirror.

These slices in the shape of rectangular parallel¬ epipeds are superposed, i.e. arranged successively, whereby a body which approximately resembles the object is formed. If the object is, for example, a potato, the volume of the potato is of course smaller than this body shaped as a rectangular parallelepiped. The produced volume of this approximate body is multi- plied preferably by a shape factor which is specific for each type of object. For a potato which resembles an ellipsoid, it can be advantageous to multiply the volume by a shape factor of about 80%, so as to resemble a correct volume. Also after multiplication by a calibrated density factor, the computer can calculate the weight of the objects. The potato density can vary to a considerable extent, and it is convenient to calibrate the density factor in known manner on the basis of a number of random samples.

The computer further calculates the maximal pro- jected length which can differ from the length seen by an observer from the side of the belt, in accordance with the discussion of Figs 1a-1c. The computer then also calculates the maximal projected width which is perpendicular to the length, and the height maximally projected from the side.

Consumers place most different demands on e.g. potatoes. Restaurant kitchens want small and round potatoes, whereas manufacturers of crisps may find it desirable to have long and thick potatoes. It is then preferred that the computer calculates an "eccentri¬ city factor" which to some extent matches the consumer's visual experience. This ' is a distinctive feature which most markedly distinguishes the present invention from the present day mechanical sorting devices which are intended for potatoes and in which e.g. a screen is used. The problems associated with these devices have been mentioned above.

The computer also calculates where on the belt the object is positioned, i.e. where the object begins and ends as seen in the longitudinal direction of the belt, and where the object begins and ends as seen in the transverse direction of the belt. Also a geometric average value is here calculated which for an ideal symmetric body would correspond to the centre of mass. These values are used for calculating control signals for a sorting means 115.

Furthermore, the computer adapts the sorting to be e.g. screen sorting to fulfil the sorting criteria which are set for potatoes. Of course, any sorting specification corresponding to certain individual requirements can be fed to the computer.

Summing up, the preferred scanning device according to the present invention is used in the following manner. One lot of potatoes is loaded on two parallel belts which are guided through the scanning device. The rotating polygonal mirror scans a laser beam across the potatoes such that the potatoes and the belts are scanned once for each millimeter of the belt. The reflected light is recognised by the detector which converts the intensity into computerisable signals. The computer emits control signals to the sorting device which is mounted downstream of the belt, where-

■ upon the potatoes can be sorted out according to suitable criteria into definite lots.

Simultaneously as the measuring of size, shape etc is effected, the object can also be identified. On the basis of the reflection curve which is specific for each object, an object can be identified. It is generally known that for example a potato absorbs at a light wavelength of 1400 nm, whereas for example a stone which may resemble a potato in appearance, reflects at this wavelength.

The identification of an object can be effected in several ways. In the preferred embodiment, a further lamp is mounted in the light-emitting means 8. The reflected light is divided into two separate wavebands and detected. The intensity of each waveband is compared with the spectrum of reflection. In the simpliest embodiment, the intensities are divided, and when going beyond or below a certain value of the quotient, the identity of the object is indicated. In the computer, the shape and identity of the object are compared. A number of different criteria of selection then control the production of the control signal for the sorting device. In the preferred embodi¬ ment, objects are sorted out which cannot, to a predeter- mined certain degree, be established as potatoes.

Of course, one and the same source of light, for example a halogen lamp, can be used both for determining the shape and for identification.

The expert can, of course, design alternative embodiments within the scope of the present invention.

Claims

1. A method for contactless scanning of an object (1) positioned adjacent or on a surface (2, 16), said method comprising: emitting light in the form of a light beam towards said object, scanning said light beam across the object ( 1 ) , detecting the light after reflection by said object (1) and said'.surface (2, 16), determining the intensity of the detected light for determining the contour of said object, c h a r ¬ a c t e r i s e d in that said object is moved relative to the source of the light beam; that said light beam in the form of a narrow light beam is scanned across the object from at least two different directions; that the reflected light is detected in at least two different directions; and that the intensity of the detected, reflected light is determined for determining the contour of the object for at least two different directions.

2. The method as claimed in claim 1 , c h a r ¬ a c t e r i s e d in that light is reflected by said object (1), a first surface (2) included in said surface (2, 16), and at least one second surface (16) forming an angle with said first surface (2).

3. The method as claimed in claim 1 or 2, c h a r ¬ a c t e r i s e d in that said object ( 1 ) is moved at an angle to the scanning direction.

4. The method as claimed in any one of claims 1-3, c h a r a c t e r i s e d in that the detected light is converted into a digital pulse train whose changes in level represent the contour of the object, low intensities of light in the reflected light giving low levels and high intensities giving high levels; that the pulse train is compared with a predetermined pattern of pulses, such that certain sequences of pulses are converted into blocks which represent distances between two points on the contour of a sectional area of the object; and that a plurality of blocks each corresponding to at least one distance for a sectional area, are registered for reproduction of the object, the repro¬ duction giving the contour of said object in the manner in which the contour is detected from the above-mentioned two different directions.

5. The method as claimed in any one of claims 1-4, c h a r a c t e r i s e d in that the light from the source of light and/or at least one second source of light in per se known manner is divided into at least two wave length-de¬ pendent luminous fluxes the intensity of which is converted into signals which are compared with a spectrum of reflection which is predetermined for said object, for identification of said object.

6. The method as claimed in any one of claims 1-5, c h a r a c t e r i s e d in that the emitted light beam and the reflected light are made substantially parallel for eliminating parallax errors when detecting.

7. A device for contactless scanning of an object (1) positioned adjacent or on a surface (2, 16), said device comprising a light-emitting means (8) which emits light in form of a light beam towards said object (1) and said surface (2, 16), a distributing means (11, 12, 13) which scans the light beam across said object and said surface, a detecting means (9) which detects the light reflected by said object and said surface (2, 16), and an electronic and computer circuit (17) which is adapted to determine the contour of said object by means of the intensity of the light detected by the detecting means (9), c h a r a c ¬ t e r i s e d in that said distributing means (11, 12, 13) is adapted to divide the light beam into at least two light beams forming an angle with each other and adapted to scan said object from two different directions; that at least one second surface (16) included in said surface {2, 16) forms an angle with a first surface (1) included in said surface (2_r 16); that said detecting means (9) is adapted to detect the reflecting light in at least two different directions; and that said electronic and computer circuit (17) is adapted to determine the contour of said object for at least two directions.

8. The device as claimed in claim 7, c h a r ¬ a c t e r i s e d in that said detecting means (9) is adapted to detect the light which is reflected back in the same ray path as the emitted light; and that said distributing means (11, 12, 13) receives and directs the light reflected by said object (1) and said surface (2, 16) and emits said light to the detecting means (9) .

9. The device as claimed in claims 7-8, c h a r ¬ a c t e r i s e d in that said distributing means (11, 12, 13) is adapted to divide the light beam emitted from the light-emitting means (8), into two light beams for scanning two different objects ( 1 ) positioned adjacent different surfaces.

10. The device as claimed in claim 9, c h a r ¬ a c t e r i s e d in that said distributing means (11, 12, 13) comprises a rotating polygonal mirror (11) which emits the light symmetrically in two opposite directions, such that each of the two separate objects (1) are scanned alternately and in the same manner, light reflected from the objects and the surfaces being received by the polygonal mirror (11) alter¬ nately and in the same manner for passing it on to said detecting means (9) .