WO1986006473A1

WO1986006473A1 - A method relating to three dimensional measurement of objects

Info

Publication number: WO1986006473A1
Application number: PCT/NO1986/000034
Authority: WO
Inventors: Jon Claesson; Jack Roald HO^/IFO^/DT; Einar SO^/RENSEN
Original assignee: Jon Claesson; Hoeifoedt Jack Roald; Soerensen Einar
Priority date: 1985-04-30
Filing date: 1986-04-30
Publication date: 1986-11-06
Also published as: DK165763C; NO168851C; FI82984B; DK634586D0; NO865009L; EP0220264A1; FI865322A0; US4773029A; FI82984C; DE3688823T2; DE220264T1; CA1253620A; DK634586A; FI865322A; ATE92617T1; EP0220264B1; DK165763B; JPS62502707A; NO168851B; DE3688823D1

Abstract

A method for three dimensional measurement of objects (1) being conveyed through or moved relative to a measurement frame (4) for, by incremental measurements of height, width and length of the object to calculate cubical and circumscribed volume of the object. Each transmitter/receiver element pair of the measurement frame (4) is calibrated by controlling the amount of the light emitted from the transmitter and the threshold level of the receiver, such that the signal/noise ratio of the receiver lies above a specific limit, simultaneously with the illumination time being kept the shortest possible. By incremental measurements at least one pair consisting of transmitter element and directly opposite receiver element in the measurement frame (4) is activated at time, and an external signal being a function of the movement of the object relative to the measurement frame is compared with the registered signals from the receiver elements for calculation of the said volumes. The measurement frame is constructed in the form of horizontal and vertical transmitter- and receiver modules (6.1, 6.2; 8.1, 8.2; 7.1, 7.2; 9.1, 9.2) consisting of said transmitter elements and receiver elements, respectively. A computer is connected to the measurement frame for calibration of the respective transmitter- and receiver elements of the measurement frame (4) and for calculation of said volumes being derived as a result of shadow images of the object on the receiver elements. A display (22) is associated with the computer for display of the volume magnitude of the object. Utilization for e.g. calculation of freight charges of objects.

Description

A METHOD RELATING TO THREE DIMENSIONAL MEASUREMENT OF OBJECTS

The present invention relates to a method for three dimensional measurement of objects being conveyed through a measurement frame for calculating the volume of the object by means of incremental measure¬ ments of height, width and length of the object. By volume is here considered both cubical volume and circumscribed volume.

A system is known from US Patent 3,513,444 for determination of volume of any object moving along a given path. The calculation of volume is effected by measuring height and^' width for each incremental distance which the object advances in the direction of conveyance. Upon each measurement there is thus calculated an incremental volume. A signal is provided for indicating that the object has moved the said incremental distance in the direction of conveyance when a pulse generator, which may be connected to the drive system of the conveyer has provided a specific number of pulses being counted by a counter circuit. The total volume will thus be given - by the sum of the incremental volume measurements. This known system makes use of a number of parallell light rays in horizontal and vertical directions. Together the light rays form a right angled matrix. Opposite to the horizontal and vertical light sources, respectively, are located the same number of photocells. That number of photocells in the light matrix not receiving light, will thereby determine the height and width of the object at the actual location. The calculated volume is according to the patent the real volume, whereas within the : transport terminology one operates with cubical volume, defined by the product of the greatest height, length and width. The indicated solution does not provide storage of sufficient number of data for making it possible to calculate the cubical volume of the object in anyone orientation thereof. The solution indicated by the US Patent has further as substantial limitation, that one dimension of that object which is to be measured always must be moved parallell to the direction of conveyance. This results in substantial limitation with respect to utilization. US Patent 3,588,480 and US Patent 3,436,968 disclose a system for retrieving data being representative of the volume of an object moving along a specified direction. For measurement of the dimensions of the object across the direction of movement, there is used a larger number of light emitters being mounted to form a matrix of parallell light rays. The dimension of the object in the direction of movement is determined by measuring its velocity and the time it is present within the light matrix. Constant light is emitted the whole time from the light matrix. According to the said patents the systems thus described are able to measure both the real and the cubical volume of the object. The cubical volume is derived by multiplication of the maximum linear dimensions in each direction of the object. In order to enable this in a simple manner with the described electronics, it is required that the object is orien¬ tated with its largest length/ width parallell to/across the direction of movement. There is indicated that the orientation of the object on the conveyor is immaterial per se, however, without indicating how the said volumes in such case are calculated.

Swedish Patent No. 425,126 discloses a method for the detection of the presence of an object within a measurement area, and measurement of its cross section in the measurement plane. For accurate measurement of dimensions is assumed that the crossection of the object is approx¬ imately circular, e.g. a log. As measurement apparatus there is described a device consisting of two arrays each having a number of transmitters and receivers. The measurement area is defined as the plane between the two arrays. Each array consists of a number of transmitters and a smaller number of receivers. If the distance between the receivers is D_ and there are the number s_ of transmitters between each receiver,..it is possible to obtain a measurement accuracy of ± D/2 s. By means of a described dual receiver, the accuracy is claimed to increase to ± D/4 s. The disclosed geometry with larger distance between the receivers than between the transmitters renders the measurement system completely unsuitable for objects having a crossection departing substantially from that of a circle.

The present invention aims at providing an improved method for three dimensional measurement, rendering it is made possible to calculate there from both cubical volume and circumscribed volume, irrespective of the orientation of the object on the conveyor belt. The present invention aims in particular to provide a solution whereby it is enabled an auto¬ matic adaptive calibration of the measurement system, automatic error detection and signal processing with improved measurement accuracy. In addition, the invention aims at using system components having relative modest requirements to tolerance values.

As a further object of the invention there is aimed at providing modular construction of the measurement frame, in order that the present invention enables simple "tailor -made'¹ solutions for dimension measure¬ ment of objects within the different volume ranges. Such modular con¬ struction will also facilitate trouble -shooting and repairs.

The characterizing features of the method according to the invention will appear from the subsequent patent claims as well as from the subsequent description with reference to the enclosed drawings.

It is readily understood that the subsequent description is only meant to illustrate non- limitative examples of the invention.

Fig. 1 illustrates schematically the measurement device according to the invention, with measurement frame and conveyor belt.

Fig. 2 illustrates a first calibration method for the measurement device.

Fig. 3 illustrates, as a non-limitative example, the measurement frame

^•> being constructed of a total of 8 modules.

Fig. 4 shows, partly in block diagram form, partly in circuit form the device according to the present invention.

Figs. 5a and 5b disclose in further detail the transmitter and receiver part of the measurement frame of fig. 4. Fig. 6 illustrates in simplified block diagram an alternative solution for the transmitter and receiver part of the measurement frame, according to the invention for further elucidation of the principle of self -configuration of the measurement frame.

Figs. 7, 8, 9, 10 and 11 disclose circuit details in connection with the explanation of the principle of self- configuration of the measurement frame.

Fig. 12 illustrates the control unit of fig. 6 in further detail.

Fig. 13 illustrates in block schematic form a receiver module in the measurement frame for carrying out synchronous modulation and demodulation.

Figs. 14 and 15 illustrates the filter characteristic of the circuit depicted in fig. 13, fig. 14 illustrating the frequency characteristic and fig. 15 the phase characteristic.

Fig. 16 illustrates in . simplified block circuit form distribution of the modulation frequency.

Fig. 17 illustrates in figs. 17a and b, respectively, the current charac¬ teristic and the light characteristic of an emitter diode, whereas figs. 17c and d illustrate modulation signal and light signal, respectively.

Figs. 18a and 18b illustrate the signal/noise ratio in connection with a good and bad receiver diode, respectively, the receiver diode measuring the absolute value of the level of light at the receiver location.

Figs. 19a and 19b illustrate for a good and bad receiver diode, respect¬ ively, the preferred solution where integrated energy of the received signal with a specific frequency and phase is caused to lie above a specific fixed thres¬ hold. Fig. 20 illustrates a typical receiver module according to the invention.

Figs. 21 and 22 illustrates in plane view and cross-section, respec- tively, an increment of the volume to be measured.

The present invention aims at providing a method for measurement of volume of all types of cargo items, where all three dimensions can be measured with sufficient degree of precision to let the values be used as basis for e.g. calculation of freight charges. The technology requires that during measurement there is a relative movement between the object and the measurement system, either by having the object moving or the measurement system moving.

Basically, the device being used for carrying out the method is designed as a frame where the objects to be measured are brought therethrough with a velocity which can be measured relative to' -the frame. An example thereof is a divided conveyor belt 5 as suggested in fig. 1. The frame exhibits two transmitters and two receiver sides, 6, 8 and 7, 9 respectively. According to fig. 1 a transmitter/ receiver unit, e.g. 6, 7 will be capable of measuring one dimension of the object 1 which is to be passed through the measurement frame. According to fig. 1, the dimension of the object 1 in z-direction is measured by the transmitter/ receiver pair A, whereas the dimension in y-direction is measured by the transmitter/ receiver pair B. The third dimension, x- direction, is measured simultaneously with the transmitter/ receiver pair A and B, respectively, being read for each movement ΔX which the object has moved in the x-direction. The pairs of detectors A and B will thus register where the object starts and where it is terminated. The difference provides the dimension of the object in x-direction. The method necessitates the installation of a velocity meter (tachometer) 3 providing signal when the object 1 has moved a length ΔX relative to the measurement system.

Although it is shown in fig. 1 that the object 1 to be measured moves relative to the measurement frame 4 by means of the conveyer 5, it will for the expert in the art immediately be understood that the object 1 possibly may be stationary, while instead the measurement frame moves relative to the object.

The present invention has as a substantial feature that the device for carrying out the method can be constructed in modular fashion, , thus making the total size of the measurement frame easily adaptable to different requirements as regards dimensions of the objects to be measured. In the example shown in fig. 1 the transmitter A consists of two modules 6.1, 6.2, the receiver A of two modules 7.1, 7.2, the transmitter B of two modules 8.1, 8.2, only the latter of which is depicted in fig. 1, and receiver B of two modules 9.1, 9.2.

Each transmitter and the opposite.- receiver therefore consists of one or several modules, each module having a length L. This therefore implies that the total length of each transmitter is n x L, where n is an arbitrary integer equal to or larger than 1. The said modules may be identical for transmitter and receiver, respectively, and interconnected in the number being necessary for the measurement tasks to be carried out. A computer 2 which processes the measurement data is not required to be programmed to handle a fixed number of modules. The computer is able to transmit test pulses, and based on returned signals, it is recorded how many modules each transmitter and receiver are composed of. This enables a simple and cost efficient structure, where different requirements to measurement ranges are met simply by putting together the required number of modules for the respective receivers and transmitters.

The invention is now to be explained further with reference to fig. 3. In the transmitter modules 6.1, 6.2; 8.1, 8.2 there are installed discrete light emitters, e.g. IR-diodes having a mutual spacing d_. In the receiver modules 7.1, 7.2; 9.1, 9.2 there are installed a corresponding number of light detectors, e.g. photo transistors or photo diodes, similarly with a mutual spacing d.

By letting only one transmitter _ and a corresponding receiver _i_ be active simultaneously, it is guaranteed that measurements take place along parallell lines. In fig. 3 as an example the transmitter 6.2.4 and the corresponding receiver 7.2.4 have been selected.

Thus, from the transmitter side light energy 10 is emitted and detected on the receiver side. Objects passing through the measurement frame will partly shadow this radiation, and incremental shadow images will thus be detected on the receiver side. A receiver is only sensitive , to the light energy from the transmitter lying right across in that the respective light detectors or receivers are activated at different instants, e.g. successively, and simultaneously with the opposite transmitter- or light diode.

By the present invention is intended to provide signal processing having improved measurement accuracy.

It is essential that the light contact which is present between a single light emitter and a single light detector becomes unambiguous. This may be solved in two ways, of which a first one is illustrated in fig. 2. The basis is here that the receiver detector measures the absolute value of the light level at the receiver location. If the light level c_ lies higher than a threshold b_ calibrated individually for each light diode, then there is light contact between transmitter and receiver. However, one may experience, as illustrated, that the signal/noise ratio may be quite different dependent on the quality of the individual light emitter and the corresponding light detector, in the example shown indicated for three diodes n, n+1 and n+2.

On basis of. measured light intensity upon complete shielding and measured light intensity without shielding, there may be determined a discrimination level IT for the light emitter/light detector pair. By measurement of absolute value it is intended to obtain an automatic adaptive calibration of the measurement system. This also requires that there is a certain minimum distance between an object and a subsequent object which are to pass through the measurement frame 4. As soon as an object 1 has passed through the measurement frame 4, the system will be entered into an adaptive calibration function such that each light emitter/light detector in the respective transmitter/ receiver pair A, B is controlled and calibrated.

On basis of said measurement, there may by aid of the computer 2 be calculated a discrimination level which under normal operation is used to determine whether a receiver is shielded or not shielded. The discrimin¬ ation level IT will thus be determined by the equation:

IT = ID + k (IL-ID),

where k is a selected constant having a value 0<k<l, ID is equal to measured intensity with complete shielding, and IL is equal to measured light intensity without shielding.

This solution, even though it provides excellent results, can however result in, as mentioned, that the signal/noise ratio becomes poor, whereby in disadvantageous cases there may be- resent possibility of in¬ correct measurements.

In order to thereby calibrate the respective transmitter/ receiver pairs individually or in groups, it is, however, according to the invention preferred the solution described below. The object of calibrating light emitter/light detector pairs is of course to optimum control the signal/ - noise ratio and the total detection time for the light detector (receiver) , in order that the signal/noise ratio is kept as high as possible, while the detection time is intended to be kept as short as possible.

The threshold value on the receiver part comparator circuit, and the total illumination time (integration time) from the respective light emitter (transmitter) are used as control paramaters. The illumination time for each light emitter/light receiver pair is added to a total time T for the complete measurement frame 4, where

N

T = E t(i), i t(i) being the illumination time for light receiver no. i_and N being the total number of light emitter/light receiver pairs in the measurement frame. The lower the time T, the more often the measurement frame may be scanned, and the more rapid the object to be measured can pass through the measurement frame. Typically, it should be required that t(i) on average should be less than 20 microseconds.

In order to improve the signal/noise ratio there is according to the invention used modulated light from the transmitter diodes. The said transmitter diodes may e.g. be IR-diodes. The modulation frequency is denoted Fm. The modulation and the associated electronics contribute to reduction of both optical and electrical noise at the receiver modules.

Unwanted light (optical noise) from other light sources, such as sun light, room lighting or the like is eliminated, as the light detectors in the receiver modules are modestly sensitive to light having other modulation- frequencies than Fm. Basically, the modulation will generate electrical noise in the frame, partly due to distribution of high- frequency modulation frequency over long lines, typically 1-3 meters, partly because large amperage in the individual transmitter elements or light emitters, typically 0,5-1,0 A, are turned off and on with the same frequency.

In order to save time, there is provided synchronous modulation and demodulation of the light emitters and light receivers, respectively. The modulation frequency is distributed in principle in the measurement frame 4 to all light emitters and light receivers. At the receiver side there is used a controlled rectifier and integrator, such as indicated in fig. 13. These have four good properties, viz: that the phase -in time for the demodulator is eliminated, due to the modulation frequency being distributed, that the detected signal in the receiver elements increases linearly with time, that detected signal in the receiver elements increases linearly with light intensity of the modulated light, and that noise having other frequencies and phase relationships than the modulation signal is suppressed according to the filter characteristics of the demodulator.

The filter characteristics of the demodulator have been indicated in fig. 14 and fig. 15, respectively. From these filter characteristics it can be derived that the filter renders the receiver insensitive to noise with frequency being the double of the modulation frequency Fm. In addition the filter renders the receiver insensitive to noise with frequency equal to Fm, but with phase shift of 90 degrees.

Both of these properties are used to improve the signal/noise ratio for the respective light receivers in the receiver modules.

Instead of distributing modulation frequency Fm over long lines throughout the measurement frame, the double frequency 2 x Fm is distributed. As indicated in fig. 16, this frequency (2 x Fm) is divided by 2, geographically speaking, as close to the demodulator (fig. 13) as possible, in order to obtain shortest possible wiring of the signal frequency Fm. The transmitter diodes being used may have rise and fall time of typically 1.5 microseconds for light emission. This has been indicated in fig. 17b, in connection with the current characteristic of the transmitter diode. By applying a modulation frequency Fm = 312.5 kHz, to the diodes, the light emission from the diodes will be 90 degrees delayed relative to the electrical modulation signal, as clearly appears from figs. 17c and d. The demodulator (fig. 13) is set to maximum sensitivity for the light signal. This yields, according to the above, that the demodulator becomes insensitive to noise caused by switching of large amperages to the transmitter diodes.

The measurement frame registers light contact between a light emitter/light receiver pair by comparing the signal from a light receiver with a threshold level. This level can be set individually for each receiver element or may be common to a group thereof.

Three variables are involved during calibration:

1. The intensity of emitted, modulated light. 2. The integration time.

3. Threshold level in the receiver-electronics.

Of these three, the variable (1) will be different from diode to diode, but for one and the same diode the light intensity will be almost constant. However, ageing and dust deposit may over time reduce the intensity, but the calibration will automatically compensate therefore.

During calibration there is initially selected a threshold value securing sufficiently high signal/noise ratio for the receiver element. Thereafter the transmitter element is activated until the receiver element detects signal reaching the threshold value. It is in this connection referred to figs. 19a and 19b illustrating signal/noise ratio relative to threshold for a good and bad diode, respectively. The threshold level is reached after N(i) light pulses from transmitter element no. _ _ For normal use there is selected an integration time t(i), where:

t(i) = (N (i) + K)/Fm,

K being safety surplus in order that received signal will lie well above the threshold value when the light path is not shielded. If ∑t(i) > TMAX for a group of transmitter/receiver element pairs, the control unit 20 (see figs. 4 and 6) will automatically reduce the threshold level and thereafter carry out a new time calibration. This is repeated until ∑t(i) < TMAX. This method implies that good transmitter/ receiver element pairs use short illumination time, see fig. 19a whereas less good trans¬ mitter/receiver element pairs (see fig_* 19b) use longer illumination time in order to obtain the same integrated .signal strength. The total time which is used is kept under a set limit, TMAX, by also controlling the threshold values. Those of the receiver elements not detecting signal above the threshold value after a maximum permissible integration time *LIM» ^are defined as defects. It is in this respect referred generally to the circuit shown in fig. 20 where a matrix detector 12 associated with an analog multiplexer 13 gets signals from the receiver elements successively, the signals from the respective light detectors being applied to a controlled rectifier 14, therefrom to an integrator 15 (see fig. 13) where the detected signal appears by a comparison with signals from a buffer and interface unit 17 through a D/A-circuit 18 in a comparator 16.

From figs. 18a and 18b representing signal/noise ratio relative to threshold for good and bad diode pair, respectively, in connection with the embodiment according to figs. 4 and 5, it is clearly noted that the signal/noise ratio for a bad diode is reduced relative to what is obtained by the preferred embodiment according to fig. 19b, which refers to the embodiment shown in general in fig. 6.

In order to make use of the complete time being available, the signal is added in the integrator 15 over the available time period. As will appear from figs. 19a and 19b, the signal increases linearly with time, i.e. with a gradient being dependent on the signal strength. By this indicated solution, the set requirements to speed are satisfied in that the lock -in time of modulated systems is avoided. Further, there is provided a signal increasing linearly with signal strength and time. In addition, the contribution from noise having other frequency- and phase relation¬ ships is reduced due to the filter characteristics of the demodulator. With the present solution a weak diode pair will thus be compensated by increasing the integration time, whereby the emitted light energy is increased. A weak diode pair will therefore be able to detect with the same signal/noise spacing as a good diode pair.

The said automatic calibration, as indicated above, enables the system, according to the invention, without measures taken by an operator, to adapt to the following non-ideal conditions:

Variations in light intensity from the transmitter elements, e.g. as a result of component tolerance or difference in component type, sensitivity variations at the receiver elements, likewise due to component tolerances or component type, different response from the receiver elements as a result of the light ray from a transmitter element not being directed with intensity maximum exactly against the corresponding receiver element, utilization of transmitter elements and receiver elements of different make and qualities within the one and same measurement system, the effect of varying background illumination due to sun light, artificial lighting or other conditions arising between the measurements, and the effect of dust collection or other conditions reducing the amount of light reaching the receiver elements.

As the calibration is made frequently, i.e. between the measurement of each object, all those conditions that may alter the transmission of light from a light emitter to light receiver (e.g. dust collection) will be taken care of by the computer. Further, product variations in the respective transmitter/ receiver element pairs will automatically be taken care of.

The adaptive adaptation also results in that precise alignment of each transmitter/ receiver pair of the module is not necessary, as long as received amount of light from the transmitter elements is not too small, i.e. lies within the threshold value.

A further characteristic of the invention is that, if required, it may be determined whether it is the transmitter element or the receiver element which is defect, and such , determination can likewise be made in connection with the adaptive adaptation by making, in addition to the normal calibration scanning (where receiver element no. _i_ detects light from transmitter element no. _i) a scanning where the receiver element and transmitter elements are mutually phase shifted, e.g. where receiver element no. i+1 or _i-l detects light from transmitter element

This automatic fault detection, which may be incorporated in the adaptive adaptation of the measurement system is initiated by a separate algorithm in connection with the computer 2 controlling that all tran¬ smitter elements/ receiver elements function satisfactory. Possible faults can be localized accurately to the actual transmitter- or receiver element.

During a control measurement a transmitter element ji__is lighted and the opposite receiver element j_ is read. It is carried out with all of the transmitter/ receiver element pairs of the corresponding modules. If the first control measurement provides as a result that one or several receiver elements are not detecting sufficient light, this will accordingly indicate fault either on a transmitter element i_ or corresponding receiver element j.

In order to determine whether the fault is caused by defect transmitter - or receiver element, a new control measurement is made. When the transmitter element i_ is activated, the receiver element j is read, the receiver element j_ is the closest receiver which with certainty functioned during the first control measurement. Due to dispersion of light, the receiver element j_ will also receive light from the transmitter element j if it is lighted, as indicated by the reference numeral 11 in fig. 3. In the selected example, J__may thus be denoting receiver element

If the receiver element j_ detects sufficient light, the system will thus be able to report that receiver element i_is defect. If, however, receiver element j_ does not detect sufficient light when transmitter element _i supposedly is lighted, the system will accordingly report that the trans¬ mitter element _ is defect. .

This self -test function vil thus be of great value with regard to simple repare and maintenance of the measurement system.

By the adaptive adaptation, the problems which are a result of varying background illumination will be eliminated.

The present device is now to be .explained further with reference to fig. 14 and figs. 5a and 5b, figs. 5a and 5b having to be viewed together for a complete understanding of the operation of the measurement frame.

As with the previous examples, the horizontal transmitters consist of two modules 8.1 and 8.2, the vertical transmitters of modules 6.1 and 6.2, the horizontal receivers of modules 9.1 and 9.2, and the vertical receivers of modules 7.1 and 7.2. The computer 2, as indicated in fig. 1, consists of a main computer 19, a control computer 20 and a control interface 21. A display 22 is connected to the computer 2 through the control computer 20.

A transmitter data bus 23 and a receiver data bus 24 extends out from the control interface 21. In addition, there is supplied an input signal from the tacho generator 3 (see fig. 1) to the control interface 21. As previously mentioned the measurement system may be composed of 1 to n modules horizontally and/or vertically. The system will self-calculate the number of modules and the resolution within each module, and defined how large the measurement axis is in horizontal and vertical directions, respectively. Although it has in the present example, for the sake of simplicity, been chosen two modules for each axis, it could of course be provided a different number of modules in the two axes, but it is a presupposition that there is an equal number of transmitter and receiver modules in each axial direction.

The description below will best be understood with reference to the enlarged circuit diagrams according to figs. 5a and 5b.

Firstly, the operation of the transmitter modules is to be explained. The transmitter modules have the following control lines: Reset, Shclk, Pxst, Pxout, Modst, L .

The signal Reset is used to stop all emission of light in all of the modules, and resetting all counters and control logic in the module.

The signal Shclk (Shiftclock) is used to move one or several light -points down along the measurement axis.

The signal Pxst (Pixel start) indicates the starting point for emission of a new light point from the top of the measurement system. After the signal "Pxst" going active, a new light point will be lighted upon a first "Shclk".

The signal Pxout (Pixel Out) denotes that the last light point in a module is lighted. It is therefore used as a start indication in the next module upon the next "Shclk".

The signal Modst (Module strobe) is active each time there is a coupling of light points between the modules. By means thereof a simple counter in the control logic of the module is able to calculate the number of light points in each individual module.

The level Lm (Last module) is tested by each module in order to know whether there are several modules connected above. If the signal Lm is active, then "Pxst" is used as a start syncronization each time it occurs. The signal is held inactive by each module lying above, such that it is only the uppermost module in the.- array which causes the signal to be active.

The contact PI to the module comes from the control interface 21 or from a module below. The contact P2 is connected to the next module, provided that such is present.

The block 25 in the transmitter modules denotes a shift register, e.g. an 8 bits shift register and current amplifiers to light diodes. D1-D6 denote modules of light diodes. Pxst arrives from the control interface 21 to start a light point from the top of the system. If the contact P2- Lmin is high, it means that the module is the uppermost in the array, and Pxst goes to the input A on the shiftregister 25. Upon the next Shclk -signal from the control interface 21, the uppermost diode in the light diode module Dl will light. For each Shclk -pulse the light point will shift down until the last diode of the diode module D6 is lighted. This occurs simultaneously with Pxout being transferred to the next module below, such that the next Shclk vil cause the first diode there to light. Simultaneously, Modst will be active, such that the control interface 21 can calculate how many light points the module contains.

It is thus understood that it is primarily tested how many transmitter modules the system has totally, whereafter the light diodes are lighted in succession from the last (logical first) module in the array and in direction towards the first (logical last) module in the array, i.e. that being circuitwise closest to the control interface 21.

In the subsequent there is to be further described the operation of the receiver modules.

The signal Reset is used to stop all measurement of light in all modules and to reset all counters and control logic.

The signal Shclk (Shiftclock) is used to move one or several measure¬ ment points down along the measurement axis.

The signal Pxst (Pixel start) denotes the starting point for measurement of a new light point from the top of the measurement system. After "Pxst" going active, a new light point will be lighted at the first "Shclk".

The signal Pxout (Pixel out) denotes that the last light point of a module is measured. It is therefore used as a start indicater to the next module upon the next "Shclk".

The signal Modst (Module strobe) is active each time there is a coupling of measurement points between modules. By means thereof a single counter in the control logic can calculate the number of measurement points in each individual module.

The level Lm (Last module) is tested by each module in order to know whether there are several modules connected above in the array. If the signal is high, i.e. active, "Pxst" will be used as start syncronization each time it occurs. The signal is kept inactive by each module laying above, such that it is only the uppermost module in the array which will cause the said signal to be active.

The signal Cbarl (Change bar 1) is used to connect Pxout from the last vertical receiver module to the start of the first horizontal measurement point. The signal is cross -coupled at the transition between horizontal and vertical modules, as clearly indicated in figs. 4 and 5a, b. The signal is used also by the control interface to define the termination of vertical measurement axis, "bar" denotes here and below a module array in one dimension, e.g. horizontal or vertical direction.

The signal Cbar2 (Change bar 2) is used to connect "Pxout" from the last horisontal receiver module back to the control interface to define the number of modules, being arranged in a horizontal direction. The signal is cross -coupled at the transition between horisontal and vertical modules.

The frequency 10 MHz is used by the receiver module internal unit 26 in which is located control logic and an internal counter to determine the read-off times for the A/D-converter connected to the light receiver elements.

Data bus part 27 of data bus 24 (D0-D7) is used for transferring digitalized measurement values from the receiver modules te the control interface 21 and the control computer 20.

The signal Cc (Conversion Complete) indicates that measurement of light point is ready to be read by the control interface 21, i.e. that the A/D- converter has sampled the light receiver elements. If several modules have active digitalization simultaneously, the control logic 26 of the receiver module will time-multiplex the measurement value to the data bus in priority from the top of the receiver array and downward.

The contact PI on the receiver modules receives its inputs from the control interface 21 or from modules below. The contact P2 extends to the next module, as shown, provided that such exists. It will be suitable to arrange buffer stages associated to each module such that all signals entering into and leaving each module are amplified. A data multiplexer being included in the said unit 26 will select output to the data bus 24 either from internal data or from modules located above. The A/D- converter being included in the unit 26 converts the measurement of the light points into digital reading. In this connection it is suitable to have an amplifier and impedance matching to the analog measurement points in connection with the A/ D- conversion. In the said unit 26 is further included a photosensor multiplexer which is able to direct one measurement point at the time to the A/D-converter. The control logic of the module which further is included in the said unit 26 both controls and synchronizes all internal logics and initiates the module below in the receiver array.

In fig. 6 is depicted an alternative version of the system in fig. 4 for further elucidation of the self -configuration principle of the measurement frame, according to the invention. It is in this connection also referred to figs. 7-12.

In fig. 6 there is used, to the extent possible, the same references as in figs. 4 and 5. The control computer 20 should by suitable manipulation of transmitter/ receiver module(s) be able to determine the number of diodes per module, the number of modules per array of modules (bar), and the total number of module arrays per frame.

In this respect there is used a diode counter, as shown in fig. 7, CE denoting the counter enable input (Counter enable) , CLEAR denoting the clear-counter input, and 0-m indicating the m+1 counter outputs.

All transmitter- and receiver modules are provided with such diode counter. This selects in succession one and one diode. The counter is incremented by the signal "Shclk" being common to all transmitter- and receiver modules. After having selected k (equal to 2^m) diodes, the counter will activate the next counter bit (bit m) locking the counter and activating the counter on the next module.

The signal Txst_n being in active state starts the diode counter on this module, and the signal Txst_n+j, in active state starts the diode counter on the next module, in direction away from the control unit 20. When the signal Txst_n is inactive, the counter is zero- set and thereby inactivates the signal Txst_n+- . All of the subsequent modules (n+1, n+2) are thereby also cleared. The signals Shclk and the first Txst- signal, viz. TxstO are generated by the control unit 20. The signal Ccstr goes active each time the first diode on a module, starting with the second module is active. The control unit 20 de¬ termines the number of diodes per module by counting the number of Shclk -periods until the next Ccstr. The signal appears as indicated in fig. 8 for transmitter module n.

Ccard_n+ι is Ccard-signal from module n+1, in direction away from the control unit 20. Ccard means "change module" (Change card) .

When Txst_n+*_[ goes active or Ccard_n+- changes level, Ccard_n will change level.

At the control unit the following ' takes place, reference in this respect being directed to fig. 9 with the signal indication a. When Ccard changes level, Ccstr (which denotes Change card strobe, i.e. module- change-pulse) will be active during a Shclk -period (while a diode is active).

The signal Cbstr (denoting Change bar strobe, i.e. module array change pulse, in other words e.g. going from horizontal transmitter to vertical transmitter) goes active each time the first diode on a new transmitter, in fig. 6 denoted by module array 2 (BAR 2) is active. The control unit determines the number of modules in each module array by counting the number of Ccstr-pulses to the next Cbstr-signal inclusive. The signal appears, as indicated in fig. 10, for transmitter module n.

The signal Cbar_n+- is signal Cbar (Cbar denoting Change bar (change module array), i.e. change transmitter orientation from e.g. horizontal to vertical) from module n+1 in direction away from the control unit 20.

The last module in the respective module array (bar) is provided with a strap which in the example shown is provided by means of a switch Sw in position B. When the signal Txst_n+- goes active or Cbar_n+- changes level, Cbar_n will change level on module with switch in position B.

At the control unit the following will occur as indicated in fig. 9 by signal b. When Cbar changes level, Cbstr -- will be active during a Shclk- period. With further reference to fig. 11, the signal Lmod (L_ast module, i.e. last module) goes active after the last diode on the last module in the frame has been active. The control unit 20 determines the number of module arrays per frame by counting the number of Cbstr-pulses until the signal Lmod receives. Lmod_n+ι is Lmod from module no. n+1, in direction away from the control unit. The module being installed farthest away from the control unit 20, in the shown case of fig. 6 module 6.2, has not connected thereto an active drive unit to the input Lmod_n+ ^* and the gate will here see a high level on Lmod_n+ ^* due to the resistance R to 5V. When this module obtains active Txst_n+j, all other modules have active Txst_n+- . The signal Lmod being active will therefore go through all modules back to the control unit. The re¬ spective switch positions for the switches A and B are shown in fig. 6. For a normal module the switch is in position A, whereas for an end module, the switch is in position B.

With the present invention there is thus provided a system which in the transmitter part and the receiver part, respectively consist of identical modules, whereby the transmitter- and receiver --parts conveniently may be adapted to the actual measurement requirements.'

By the solution shown in fig. 6, the signal Txst-Jmpl (denoting trans¬ mitter start signal with a jump to module array 1) will cause counting starting from module no. 3 (6.1) to module no. 4 (6.2) , inclusive, where- after the signal Txst-Jmp2 causes the module array 2 to be scanned, from module 1 to module 2, inclusive. For the receiver part the scanning will be made successively through module array 1 and 2, i.e. 7.1, 7.2, 9.1 and 9.2. In fig. 5 the scanning, in the example shown, took place on the transmitter part from the last transmitter element in module array no. 1 to the first transmitter element in module array no. 2, and in the receiver part from the last to the first receiver element in receiver -module array no. 1, and thereafter from the last to the first receiver element in module array no. 2.

Common to both embodiments is thereby that the horizontal and vertical transmitter modules of the transmitter part signalwise are connected in series in such a manner that the scanning signal is fed directly to the logical first transmitter module, whereafter the transmitter elements in each transmitter module are selectively activated, module by module, from logical first transmitter element to logical last transmitter element in the module array, only one transmitter element and a receiver element located directly opposite in the measurement frame being activated at a time, and that the vertical receiver modules and horizon¬ tal receiver modules of the receiver part are similarly arranged in series, but signalwise such that the first module array is scanned from logical first to logical last module in the set, where after the outermost located module array is scanned from the logical first module to the logical last module of the array.

By "identical" modules in the transmitter- and receiver parts, respec¬ tively, is meant that respective modules are basically constructed equal, but can thereafter be . provided by internal rewiring facilities, as e.g. shown by str. 1-5 in fig. 6.

By the solution which is shown it is possible, dependent on the appli¬ cation to configurate the measurement system without restrictions by assembling a number of equal/unequal standard modules. The hardware of the measurement system is able to register how many sensors which are present in the various modules of the actual configuration and will be able to operate even in the event that a transmitter/ receiver module pair does not have the same number of transmitter elements and receiver elements. The system will thus register the total length and resolution both in horisontal and vertical part of the measurement system, as these parts, as mentioned suitably can be different.

In the measurement system of the dimension discussed herein, it may be possible to let several pairs of transmitter elements/ receiver elements be active simultaneously, and this can be made if the light beam 10 from a transmitter element does not cover more than one of the active receiver elements.

It is important to note that by continuous monitoring of the velocity and the position of the object, the velocity may suitably be changed, while the measurement is in progress, without this reducing the measurement accuracy. Further, the orientation of the object during the measurement is without importance for the measurement result.

By the present invention is it made possible measurement of both the cubical volume and the circumscribed volume. When the object passes the measuring plane, as indicated in figs. 21 and 22, the highest point (HMAX) will be measured at each reading based on the shadowing of the horizontal light rays. When the object passes the measuring plane, the outer right point (see fig. 22) on the object (BMAX) and the outer left point on the object (BMIN) will be measured dependent on the shadowing of the vertical light rays. The values of HMAX, BMAX, BMIN are measured and stored for each increment ΔX which the object moves in the direction of conveyance, see fig. 21.

The cubical volume VK is defined as the volume of the least right angled parallell -epiped which encloses the object. Generally, the cubical volume will be larger than the true volume. The cubical volume VK is calculated by storing the contour positions BMAX, BMIN, HMAX for all readings over the extent of the object in the direction of conveyance. When' all data are present, it will be possible by means of a suitable algorithm to- read through to four points among the amount of points BMAX and BMIN, such that each of the four points lie on a side edge each of the least rectangle enclosing the vertical projection of the object. The area of this rectangle is denoted by AR. The cubical volume of the object is thus given by VK = AR x HMAX, HMAX being, as mentioned, the greatest height of the object. The circumscribed volume VO is defind as the sum of partial volumes, vd, calculated over the extent of the object in the direction of movement. Thus, the circumscribed volume becomes:

M VO = Σ_ vd(i) i

where M is the total number of measurements of the object. The partial volume vd(i) in a measuring plane "i" is defind by:

vd(i) = h(i) .b(i) .Δx where h(i) is the height of the object in the measuring plane, b(i) is the width of the object, and ΔX is the incremental distance between each measuring plane, see figs. 21 and 22. VO is generally larger than the true volume, but they become equal if the cross -section of the object is rectangular over its entire length.

As scanning of the transmitter element and receiver element of each module is synchronous, relatively small requirements will be set to the precision of the light emitters in the transmitter modules, so that cheap components can be used, as it is basically only the direct light from the light emitter in the transmitter module to the light detector in the receiver module which is detectable. With the present measurement system it is visualized that one could reach a resolution accuracy of approximately ± 5 mm, when using a conveyance velocity less than approximately 30 m per minute. However, this indication of the accuracy of resolution should in no manner be considered as limitative to the present invention, but only express a typical example.

Claims

1.

A method for three dimensional measurement of objects to be conveyed through or moved relative to a measurement frame for, by means of incremental measurements of height, width and length of the object, to calculate the cubical or circumscribed volume of the object, for each transmitter element in the measurement frame there being provided an opposite receiver element which only detects light during the time period when the transmitter element is active, the transmitter- and receiver elements being mounted in arrays on modules with a number of elements in each transmitter- and receiver module, one or several modules being present along each measurement axis, and an external signal which is a function of the movement- of the object relative to the measurement frame being compared with the data signals derived from the receiver elements for calculating said volume, said data signals being derived on basis of shadow images on the receiver elements caused by the object, ' characterized in: that for each individual receiver element or groups of receiver elements there is determined a threshold level forming criteria for whether the receiver element is illuminated or not, that the modules, being in series, are activated by input of a start signal in the first of the modules, which signal is caused to be shifted along the interconnected modules so that at least one and one element is activated at the time, the start signal when it reaches the end of each module causing registration of the module end, when it reaches the end of each module array for the respective measurement axis causing registration of that end, and when it reaches the end of the inter¬ connected module arrays, causing registration of the non-presence of subsequent modules in the form of a "last module" -signal, that all control- and data signals for the transmitter part and for the receiver part are fed from one module to the next through the trans¬ mitter modules and receiver modules, respectively, and that said transmitter part and receiver part are controlled from a computer in which said volume is also calculated.

2.

A method according to claim 1, characterized in that each transmitter/ - receiver pair is calibrated by controlling the amount of light emitted from the transmitter, and the threshold value at the receiver, so that the signal/noise ratio for the receiver is above a specific limit, simul¬ taneously with the illumination time caused to be the shortest possible.

3.

A method according to claim 1 or 2, characterized in that the emitted amount of light from the transmitter is determined by controlling the intensity or duration of the light being emitted during the active period of the transmitter element/ receiver element.

4. A method according to claim 1, 2 or 3, characterized in that the light from each transmitter element is modulated, and that the receiver elements are made sensitive only to such modulated light.

5.

A method according to claim 1, characterized in that the mutual velocity of the object and the measurement frame is altered while the measurement is in progress.

6. A method according to one or several of the preceding claims, charac¬ terized in that the transmitter/ receiver elements of the measurement frame make use of visible or invisible light.

7.

A method according to one or several of the preceding claims, charac¬ terized in that two or several transmitter/ receiver element pairs are activated simultaneously.

8. A method according to one or several of the preceding claims, charac¬ terized in that the transmitter/ receiver modules output as a result of said start signal electrical control signals indicating the positioning in the measurement system of the respective modules and characteristics of the module, that the transmitters are made operative in succession from logical first transmitter element in first transmitter module to logical last transmitter element in last transmitter module, a first control signal being sent to the computer when the start signal reaches the end of the module, a second control signal when the start signal reaches the end of each module array for each respective measurement axis, and third control signal when the start signal reaches the end of the inner- connected module arrays, the control signals coming from the modules being characteristic of the composition, form and size of the measure¬ ment system, and the computer controlling the modules not requiring to be preprogrammed to know the geometry of the measurement system.

9. A method according to one or several of the claims 1-6 and 8, charac¬ terized in: that the horizontal and vertical transmitter modules of the transmitter part are signalwise connected in series so that the scanning signal is fed directly to logical first transmitter module in the module array, where- after the transmitter elements of each transmitter module are selectively activated module by module, from logical first transmitter element to logical last transmitter element in the module array, only one transmitter element and a receiver element in the measurement frame located right opposite being activated at a time, and that the vertical receiver modules and horizontal receiver modules of the receiver part similarly being connected in series, but signalwise such that the first module array is first scanned from the logical first to the logical last module of the array, whereafter the outermost module array is scanned from logical first module to logical last module of the array.

10.

A method according to one or several of the preceding claims, charac¬ terized in that the said external signal is supplied from a tacho generator, and that the value of the volume calculated by the computer is supplied to a display device.