Lumbering optimizing device and method of confirming wood node section

(54) (56)

PATENT HOLDER MacMillan Bloedel Ltd, Vancouver CA

INVENTOR

REPRESENTATIVE TITLE

CALLED PUBLICATIONS:

Jan Erik Aune, Vancouver CA, Terence James Arden, CA, Mary Sharlene Yap, Vancouver CA AWAPATENT AB

Timber optimization systems as well as a procedure for twig definition

Vancouver (57)

SE 348 558 G01N 33/46), SE 418 774 (G01N 33/46), SE 407 982

G01N 33/46), SE 467 940 (G01N 33/46), DE 3 034 543 (G01N 33/46)

WOOD SCIENCE Vol. 14 No. 3 Application of Automatic Image Analysis to Wood Science, Carles W. Me Millin FOREST PRODUCTS JOURNAL Jan 1987 pages 56-62

SUMMARY: _One wood optimization system and a twig-defining method in which a conveyor means (12) feeds stem blocks or boards (16) through a profile station (18). This cooperates with an analyzing means, a defining means and a saw setting means, which define a sawing solution, and with a positioning means which positions each stem block or board (16) relative to a sawing member in dependence on the sawing solution. In addition, a data acquisition station (20) comprising radiation means (26) transmitting electromagnetic radiation through each trunk block or board (16), and a radiation receiving means (32) sensing the amount of radiation passing through them are further arranged to collect data indicating the density in local areas of the stem block or board (16) and which are compiled into a density map. This is then used by the analyzer to define a sawing solution that takes into account defects, such as twigs, which can be read from the density map.

<img file="SE504547C2_D0001.tif" />

The numbers in brackets indicate international identification code, INID code. Letters in clamps indicate international document code.

504 547

Field of the Invention

The present invention relates to a wood optimizer, and more particularly, the present invention relates to a real-time scanning system for scanning, producing an image and controlling the orientation of the board and positioning saws in accordance with the analysis of the image so produced.

Background of the present invention

The idea of ​​analyzing wood using electromagnetic radiation has also been suggested long before any equipment that could successfully commercialize such a thought was available, see, for example, an article by DG Miller entitled Detection of rot in wood by electronic X-ray fluoroscopy as was published in British Columbia Lumberman in October 1964. The idea of ​​using X-rays to detect rot in wood was said to be old at this time and reference was made to a study as far back as 1929. The theory assumed that provided the rot had progressed sufficiently to significantly reduce the density of the wood, the absorption of X-ray would also be reduced. . The article concludes that X-ray illumination offers a fast, non-destructive method of conducting internal inspections of wood after root pockets and metal.

An article entitled Defect detection in the lumberstate of the art by Szymani and McDonald in the November 1981 issue of the Forest Products Journal describes the various techniques for analyzing boards and logs. In connection with X-ray copy, reference is made to the above-mentioned article by Miller and to real-time X-ray, television and X-ray cinematography so that the images can be projected directly through a television network. Also described is a system in which an X-ray sensitive

504 547 device detects the amount of X-ray passing through the object and by filtering and amplifying the signal, an indication of the occurrence or absence of defects can be obtained and the output signal is used via computer for wood classification and sawing decision. Particular attention is paid in this article to the Scintaflex system, which, in this case, directs neutrons to a point-shaped surface on the board and senses the amount of radiation that penetrates the board to obtain an indication of the density of the point.

Various techniques have been used or described to identify twigs and / or rot in timber, for example in the article Locating Knots by Industrial Tomography - A Feasibility Study by Taylor et al., Published in the Forest Product Journal in May 1984 or in the article A Computer Vision System that Analyzes CT Scans of Saw Logs by Funt and Bryant published by IEEE Computer Society Conference on data monitoring and pattern detection.

20Funt and Bryant also published an article entitled Detection of Internal Log Defects by Automatic Interpretation of Computer Tomography Images, published in January 1987 by the Forest Products Journal, which describes in detail_ the analysis of a stock's cross-sectional histogram of a density map produced by irradiation of a stock using x-rays and analyzes this histogram to determine the location of twigs and rot.

Optical scanners have also been used to determine the location of surface defects in wood by surface color differentiation.

Automatic woodworking systems (ALPS) have also been described in which the information obtained by an optical scan of the board is analyzed on the basis of tone, color, structure and pattern recognition to determine the location of surface defects in a wood piece and to use the information thus generated.

504 547 to produce a sawing solution for signing the inspected boards to obtain the optimum of useful wood from the inspected board.

An article in the IEEE Transactions on Pattern Analysis and Machine Intelligence volume TAMI-5 no. November 6, 1983 entitled Code Identifying and Locating Surface Defects in Wood - Part of An Automated Lumber Processing System by Conners et al., Describes a system in which an image is produced by optical scanning, such as by means of a laser scanner and the image is examined to assess its tonal properties, i.e., the degree of clarity, its structure or pattern qualities and pattern recognition to detect defects in the wood material and the location of these defects.

An article titled ALPS - Potential New Automated Lumber Processing System by McMillin et al, in Forest Products Journal volume 34 no.1 January 1984, deals with detection and location of defects in a log and provides an optimal partitioning solution based on this information. In this publication, a log is processed by scanning using photon tomography and computer reconstruction of axial projections from three different angles to locate defects in the timber. After cutting the timber into boards, the boards themselves are examined with video cameras and the image information is digitized and analyzed for tonal and structural qualities and an optimal partitioning strategy based on the location of the defects is executed.

The article A Prototype Software System Locating and Identifying Surface Defects in Wood by Conners, given at the Seventh International Pattern Recognition Conference in Montreal Canada July 30 - August 2, 1984 and published in Protocol Volume 1 provides further details regarding the distinction of flawless wood from defective wood in wood using optical scanning techniques in which the underlying edge, twigs and flawless wood are distinguished.

504 547

United States Patent 3,931,501, published January 6, 1976 by Barr et al., Discloses yet another scanning technique for detecting and indicating the surface defects on a piece of wood, and then a dividing pattern is specified to edge the timber into different wood elements. Canadian Patent 1,146,051 issued May 1983 by Strandberg et al discloses a system for optimization based on measuring the contours of a piece of timber and sawing based on an optimization program which positions the timber for sawing in the edge mill and regulates the edge mill accordingly.

Canadian Patent No. 1,125,148 discloses an optical sensor for detecting irregularities along the longitudinal edges of a log and controlling the position of the log for sawing and eliminating these irregularities.

It will be apparent from the above that various techniques have been proposed and in some cases created for scanning boards to create images of the surface of the board which are then analyzed, first to determine defects and types of defects and location of these surface effects and then to specify a optimal dividing solution for edging the board and positioning the same for subsequent edging. These systems primarily use optical scanning techniques to create the image to be analyzed. However, in some cases it has been suggested to use electromagnetic radiation by directing a needle tip of such radiation through the board and sensing the intensity of the radiation after it has crossed the board to get density indications.

Thus, it is known to obtain a real-time analysis of the board and to provide an optimal partitioning solution, ie to analyze the image produced by scanning the board in real time to determine the position of defects and from this to form an optimal partitioning 504 547 pattern. Such real-time analysis has been based solely on optical scanning and analysis of the image thus produced. However, it will be appreciated that optical scanning can only detect surface defects and is in many cases limited by the surface of the stem block being scanned so that the rough surface of a roughly sawn stem block cannot be treated using this technique.

X-rays or electromagnetic inspection technology have been used primarily for the analysis of whole logs rather than logs or unpaved timber and the image (s) produced have been analyzed for defects and a dividing solution for logs. It has been suggested that X-ray tomography or screening may be used to detect defects in wood. So far, the published information has been related to the use of X-rays to inspect wood and has not been of a practical nature, ie operated in real time (capable of operating at conventional production rates).

Brief description of the present invention

It is an object of the present invention to create a real-time system for scanning a stem block to create an image of the stem block based on density variations, analyze the image created, and specify a sawing solution fast enough to maintain reasonable machine production speed (real time).

Broadly speaking, the present invention relates to a timber optimization system comprising a conveying device for transporting logs or logs through said system, a profiling means for generating a signal indicating the edge of said logs, a data collection station with means directing an electromagnetic radiation line through each stem block as it is moved by said conveying device through the data collection station and means for locally sensing the amount of radiation traversing each such stem block

504 547 to thereby provide data indicating the densities in local areas of each trunk block, means forming a density map from said data, image analyzing means for analyzing the density map and signal for determining the size and location of defects on the density map, means for determining a sawing solution based on the image analysis and determination of the size and location of the defects and said signal as determined by said profiling means; means adjusting a sawing agent in accordance with the sawing solution developed for each stem block and means for adjusting a logging positioning means for placing each stem block relative to the sawing means in a position determined by the sawing solution for each individual stem block.

The sawing means may be an edge mill in which case the positioning means for the stem block angles the stem block relative to the saws in the edge mill to determine the direction of the division through the stem block or alternatively the sawing means may be an adjusting saw and the positioning means intended for the stem block will extend axially along axis of the stem block) an axial end of the stem block.

Preferably, the data collection station comprises a plurality of spaced groups of sensors which simultaneously measure the local densities of the trunk block over different regions of the trunk block so that the time needed to collect data representative of the density can be performed in a short time.

The density map is preferably formed in the smallest number of image buffers, depending on the width of the master block being processed, so that if data from more than one group of sensors can be placed in offset ratio in the same image buffer, i.e., more than twice the number of significant pixels affected for the width of the master block available at hand in a single image buffer comes more than

504 547 a length of the master block to be stored in an image buffer, thereby reducing the time required for the image analysis since the number of image buffers to be analyzed by the image analyzer is correspondingly reduced.

The functions of the data collection, the formation of the density map, the image analysis and the calculation of the sawing solution are performed asynchronously and reported to a control computer which controls the operating system according to the position of a trunk block and the state of the various functions to maximize the speed of the transport device.

Brief description of the drawings

Further features, objects and advantages will become apparent from the following detailed description of the preferred embodiments of the present invention which are described in conjunction with the accompanying drawings in which:

Figure 1 is a schematic side view of a machining system incorporating the present invention.

FIG. 2 is a schematic illustration of an array of electromagnetic radiation detectors positioned to intercept the radiation line passing through the board.

FIG. 3 is a schematic illustration of an optimization system in an edging embodying the present invention.

Fig. 4 is a schematic illustration of an optimization system in a jute saw incorporating the present invention.

Description of the preferred embodiments

As shown in Fig. 1, the scanner 10 is composed of a basic main conveyor 12 which, in the arrangement shown, is a chain conveyor with pins 14 on which the stem blocks or the uncut wood 16 are placed by suitable means (not shown). The conveyor 12 moves the stem blocks 16 by the pins

504

547 grips one side of each stem block, whereby the stem blocks are automatically positioned relative to the conveyor so that the position of each stem block is known with a gap predetermined by the gap between the pins 14.

The conveyor 12 moves the stem blocks one by one past a profiling station 18 (Figs. 1 and 3), which has a first sensor which may be any suitable sensor, such as an optical scanner adapted to determine the edge and periphery of the board. Such devices are well known, for example, the optical sensor sold by Lloyd / Softac in Vancouver, BC under the trade designation LS-8600.

After passing through the optical scanner or sensor 18, the trunk block 5 passes into a data acquisition station 20 comprising an X-ray or a suitable density sensor which will determine the local densities of the trunk block 16.

Preferably, the density sensor will comprise at least one source 26 of X-rays adjustable by a collimator, generally designated 22, to reduce the emitted electromagnetic radiation to a flattened solar spring form, as illustrated at 24, to obtain a narrow line of X-rays as strain block. In Figure 2, four laterally spaced X-ray sources 26 are shown, each projecting a line toward the stem block passing through the scanner 20. The lines from each source 26 are arranged substantially end to end in a substantially axially straight line so as to extend substantially the entire length of the stem block. If desired, the lines need not be in axially straight line. Adjacent ends of these lines may overlap somewhat, but in that case the detectors 32 affected by the overlap must be calibrated accordingly. It is also possible to have small spaces between adjacent ends of these lines radiating from nearby

504 547 which will leave unexamined portions across the stem block, for example in the area of ​​the transport chains 12.

Electromagnetic radiation is enclosed within a housing 28 on scanner 20 by a conductive lining near the bottom edge, i.e., where the stem blocks pass through with a flexible conductive lining curtain at the incoming and outgoing side as indicated at 30. The curtain 30 can be folded away by the stem block which enters and leaving the detector 20.

The spring-shaped flow of photons 24 from each of the ends 26 described above forms a line of photons extending substantially over the entire length of the stem block 16. The line is directed to a side-by-side array of X-ray detectors 32 composed of a plurality of side-by-side scintillators 33 and corresponding photodiodes 34. These photodiodes 34 are each of substantially the same width and are uniformly arranged along the length of the photons line. The width of these diodes is determined by their number placed in a side-to-side ratio along the length of the photon line. Generally, there will be at least 4 (four) such diodes per inch to obtain a usable image and normally no more than about 16 (sixteen) since under normal use it is of little practical advantage to reduce the width of the detector area below it. The applicant has found that the use of ten diodes per inch offers a very satisfactory resolution.

In the device of Fig. 2, various diodes are designated by reference number 34 (only some of the diodes are designated) and these diodes are connected by a printed circuit device or the like with an analog to digital converter generally designated 36 which in turn is connected to an output line. generally designated 38 to direct the digitized data collected through each group 35 of array 32 to a suitable computer 62 as will

504 547 is described below.

Each diode has a specific address and their output voltage is read and digitized sequentially by group. Preferably, the diodes are divided into special groups as indicated at 35. In the arrangement shown, a group is intended for each of the transmitters 26 but this is not necessary. The diode outputs of each group 35 are read and digitized sequentially, all the groups 35 being read and digitized simultaneously. By thus dividing the diodes into groups along the radiation line transmitted to the trunk block, the time required to collect the relevant data along the entire length of the trunk block is reduced.

For example, if four separate groups 35 are arranged and the outputs of the diodes in each group are read simultaneously and digitized, only a quarter of the time is needed to sequentially read and digitize the outputs of the entire line (assuming each group 35 contains the same number of diodes). Thus, the diodes will be divided into a suitable number of groups to ensure that the necessary data is collected within the required time. In a particular example of the present invention, the diode outputs are read and digitized sequentially in groups of five hundred and twelve diodes at a frequency of 100 kilohertz as the trunk block passes the sensors.

In the embodiment according to Figs. 1 and 3 (Edge Optimizer), the stem blocks pass, after being sensed or scanned by scanner 20, along the conveyor 12 over the open portion 39 and then to the second conveyor 40 (see Figures 1 and 3) which is driven in the direction indicated by arrow 41 and moves the stem block from the conveyor 12 to the positioning station 42 wherein the leading edge of the stem block 16a of FIG. 1 is moved to a position against adjustable stop pins 44 which are movable as shown by arrows 46 to be aligned with the leading edge of the

504 547 stem block having the desired angle relative to the edge conveyor feed conveyor 48. The conveyor 48 is driven by a motor 50 intended for the edge mill. The many rollers, such as those shown schematically at 52, are located on opposite sides of the conveyor 48 which is relatively narrow relative to the the width of the stem block perpendicular to the direction of travel of the conveyor 40 so that the stem block cannot tip when conveyed by the conveyor 48. In general, a retainer (not shown) will cooperate with the top side of each stem block as it is transported into the edge mill 54.

The conveyor 12, which transports the stem blocks through the two sensor stations 18 and 20, is driven by a suitable feed motor 56, while the forward conveyor 40 is driven in the direction of the arrow 41 by a feed motor 58.

The main data collection, analysis and control systems for the embodiment of Figures 1 and 3 are schematically shown in Figure 3.

In the embodiment shown in Fig. 3, the surface profile of the board or trunk block is obtained from profiling station 18 and fed into profiling computer 60 via the line and data for density mapping is obtained from data acquisition station 20 and fed into density mapping computer 62 via communication line 38.

The image analysis computer 64 receives data from the computer via line 66 and from computer 60 via lines 68 and 70. The surface profile information from computer 60 is superimposed on the density map of computer 62 to increase the accuracy of the image analysis for defect classification.

The image analysis information which determines the location of defects including twigs, rot and board dimensions, etc., is then fed to the sawing solution computer 72 via line 74 and used to determine the sawing solution. The determination of the sawing solution also uses signals from the computer 60 which enter via the line 68.

504 547

The aforementioned data collection and computers are all operated asynchronously and their states are reported to a main processing 84 computer which is responsible for the overall control of the process. The process controller computer 84 synchronizes the function of all the components of the system and therefore has inputs from computers 62, 64 and 67 as well as from the wood detectors 82 (only one shown) which reads the presence of a trunk block 16 being introduced to the equipment . Information from detector 82 is fed via line 86 to process controller computer 84. If desired, other detectors may be present at other stations, such as one at each station 18 and 20, section 39, conveyor 40 and position 42.

Each of the computers 60, 62, 64 and 67 is connected to the process controller computer 84 through lines 94, 96, 98 and 100 for transmitting data therebetween. The processing computer 84 controls the operation of the various conveyor motors for the conveyor via the conduit 88. The main supply motor 56 in the conveyor 12 is connected to the conduit 88 via the branch conduit 90, the supply motor 58 in the conveyor 40 via the branch conduit 92 and the edge plant's feed motor 50 via the branch conduit 93.

The computer 84 includes computer capacity 76 for controlling the stem block positioner 42 on the basis of the input from the various computers including the sawing solution computer 72 and transmits the sawing solution information to the positioner 42 via line 80 and edge network setting 54 via line 78.

In operation of the embodiment according to Figs. 1 and 3, the first stem block or board 16, designated at 16b in Fig. 3, is supplied to the conveyor 12 at the intake end thereof and is moved perpendicular to its longitudinal direction in the direction of arrow 102 first to the optical scanner 18 where profile is examined so that the edges of the stem block are detected and their location determined. The stock block is continuously fed through the scanner 18 at a predetermined feed rate

504 547 of the feed motor 56 so that the dimensions and profile of the stem block can be easily determined. The stem block then passes at the same feed rate through the data collection station or scanner 20 where X-rays pass through the stem block and a density profile is measured by the row of detectors 32 in a continuous manner as the stem block moves past it.

The data collected in scanner 18 is transmitted through a line 61 to the data acquisition or profiling computer 60 wherein the profile of the trunk block is determined. The information from the scanner 20 is transmitted via line 38 to the data acquisition and density mapping computer 62 wherein a density map of the density profile of the scanned stem block is obtained. The density profile is based on the signal strength per pixel of the created image which, in turn, through the signal attenuation of the X-rays passing through the board and received at a given time period by each of the various detectors in the line 32. Normally, the signal will be inverted so that the regions with the largest density will appear in the image as the areas with the highest intensity on the grayscale in the image analysis.

The time for image analysis can be significantly reduced if the images from more than one scanner link (equal to the number of pixels across a standard image buffer) can be accommodated on a single image buffer. This can be accomplished by using information from the edge scanner 18 which determines the maximum width of the board so that if the number of pixels across this maximum width is less than half the height of a pixel buffer, two scan lengths can be accommodated in the same buffer. If the number of pixels across the maximum width of the master block is less than one-third of the number of pixels at the height of a buffer, three scan lengths can be accommodated in a single buffer, etc.

By packing more than one scan length into one

504

547 individual image buffer reduces processing time for analysis of all images. The number of buffers that must be analyzed and processed can in many cases be reduced to half and in some cases to one third, thereby reducing the analysis time so that the feed rate (conveyor 12) can be increased.

A histogram of the resulting intensities of the pixels in the grayscale image (sensed densities) is produced by accumulating buffer histograms and the computer 64 analyzes the total histograms of the intensities (densities) corresponding to the density of good wood, the density of red and low density prepared voids such as holes or dry rot. For example, if relatively dry wood is treated, the density of rot is markedly less than that of the good wood and it can be determined on the basis of its low density. In wet wood (i.e. wet wood may have been transported via water), however, the density of rot may well be higher than the density of the good wood and the particular density spectrum depicting rot will be higher than that of the good wood.

High-density ranges can also be found by analyzing frequency information in the image.

The applicant has discovered that twigs can be determined by sudden changes in density, ie a relatively sharp interface between the twig area with its high density and the lower density of the area with good wood. It has been discovered that twigs can be recognized in this way by using a one-dimensional part-2-G digital filter, one dimension of which is substantially parallel to the longitudinal axis of the stem block to determine the sudden changes in density as the scanner arrives and leaves the twig. This technology has been discovered able to pinpoint twigs accurately and quickly without the need for unnecessary computer capacity.

In any event, data is analyzed to determine which pixels represent twigs, which pixels

504 547 which represents good wood and which pixels represent red. The relative positions of these pixels are known so that the position of red and twigs in the board can be determined.

On the basis of this information plus information collected in the detector 18 and processed in the computer 60 which determines the outer edges and edge of the stem block, all defects located near the edge of the stem block can be eliminated and the work of the computer 64 simplified as the outer contours of the board are determined by the computer 60.

The information from the computer 64 and the information from the computer 60, described above, are fed to the sawing solution computer 72 which, based on the correct values ​​for the grading of the wood products, is programmed to optimize the value of the material which can be sawn from the stem block in terms of quality and volume so that the resulting sawing operation can provide a maximum yield. The computer 72 supplies the sawing solution to the computer 84 which positions the edge saws of the edge mill via the edge mill setting 54 and calculates in the calculation section 76 the position of the sprinters 44 and commands movement of the sprinters 44 in the positioner 42 accordingly.

The main or control computer 84 controls the function and speed of the system. Each of the computers 60, 62, 64 and 72 feeds information to the control computer 84 and gives signals to the computer 84 as they have completed their tasks for each master block as it has passed through the system. The speed of the conveyor 12 is controlled by the motor 56 which is controlled by the computer 84 for normal movement at maximum speed to move the stem block through the system to the edge saw 54 as quickly as possible and to fit the following stem block into the system. If the calculations of the computers are not completed for a particular trunk block when the trunk block enters the conveyor 40, the computer 84 slows down or

504

547 stops the conveyor 12 until all operations for that stem block are completed and then increases the speed of the conveyor 12. This ensures that the line operates at a maximum speed except when an unnecessarily complicated operation is encountered which requires significantly more time than average. If this occurs, the line speed (conveyor 12) is lowered or stopped depending on the particular control use.

When it comes to positioner 42, the stem block 16b has been fully analyzed by the computers 60, 62, 64 and 72 and the angle of the stem block towards the input conveyor 48 has been determined. In order to position the stem block in the correct orientation, the pins 44 are adjusted (only two are shown but more can be used depending on the length of the stem block to be positioned) as indicated by arrows 46 and the stem block is advanced to positioner 42 of the conveyors 40 driven by the motor 58 which is activated. when a stem block is in position to move the edge of the stem block against the pins. The conveyor 40 can simply be a conveyor of a type with a flat belt that can slip relative to the stem block so that as the conveyor 40 pushes the stem block toward the most protruding of the pins 44 so that the end of the stem block slides on the conveyor and the stem block is shifted to the correct orientation in its longitudinal axis with the direction of travel of the conveyor 48 toward the edge 54 as indicated by arrow 49.

When the edge of the stem block has been positioned against the pins 44, the feed motor 50 is activated to the edge mill to move the stem block in the direction of arrow 49 through the edge mill 54 to saw the stem block as required.

The above description is intended for edge milling systems, however, it may as well be applied to a jigsaw saw system such as that shown in Fig. 4.

In Figure 4, the same reference number has been used for

504 547 to designate the same parts of the invention described above and these elements will not be described again.

It should be noted that in the embodiment of Fig. 4, no profiling scanner 18 is included. In this case, the data generated by the scanner 20 is transmitted via a line 61a to a data acquisition and board profile computer 60a which determines the edges of the board on the basis of a noticeable change in density formed as the edge of the board crosses the radiation line and a density peak is formed substantially along the line. formed where the edge of the edge crosses the plane forming a surface of the stem block. Thus, the computer 60a determines the position of the edge of the stem block as well as the line where the edge edge crosses the board or stem block surface. This technique of determining the edges of the edge edges of the stem block cannot determine the slope of the edge edge as can be done with an optical sensor such as that shown in Figures 1 and 3 (sensor 18), thus the sensor 18 can be avoided and the board profile is determined as described immediately above. when using the computer 60a. It is preferred to use an optical scanning system using the optical scanner 18 and the computer 60 as these allow for more accurate determination of a sawing solution, i.e. if the slope of the vane edge is relatively flat near its transition to the flat surface of the stem block, it may be acceptable to bring more of the edge into the sawn boards.

Optical scanner 18 and computer 60 may be used in the embodiment of Fig. 4 instead of computer 60a or in the embodiment of Fig. 3, computer 60a may be used in place of optical scanner 18 and computer 60.

In the embodiment of Fig. 4, the conveyor 12 is unloaded on an input conveyor 200 carrying the stem blocks (hereinafter referred to as boards) into the trim saw 202 having settings 204. Thus, the embodiment of Fig. 4 can be applied to the conveyor 12 and the sensor 20 is sensed by a sensor 20. similar to the stem blocks fed to the conveyor 12 in the embodiment of Fig. 3.

The boards conveyed through the sensing station 20 and along the conveyor 12 are sensed and the sawing solution is determined in substantially the same manner as described above in Fig. 3. All this information is fed through the process controller computer 84.

In this case, the computer capacity 76a for positioning the board is utilized for control via line 206. The position of the end plate 208 can be moved in and out as shown by arrow 210 to control the function of positioning motor 212 via line 214. Positioning motor 212 drives the interconnected rollers 216 to forcing the longitudinal end edges at a longitudinal end of a board positioned on rollers 216 to abut against plate 208 thereby positioning one end of the board.

Simply put, as soon as the sawing solution is known, it is necessary to position the board longitudinally relative to the jigsaw 202 and to adjust the jigsaw saw 204 according to the sawing solution, which is accomplished by the process controller computer 84 and its connection line 218.

The jigsaw saw conveyor 200 is driven by a feed motor 220 which is controlled from the processing computer 84 via line 88 and branch line 222.

In operating the system shown in Fig. 4, the board of conveyor 200 is moved to a position on rollers 216 as soon as the sawing solution has been promoted to process controller 84. Plate 208 will normally have been placed by the positioning computer for boards in section 76a before the board reaches conveyor rollers 216. With the plate 208 positioned as required, the rollers 216 are actuated by the positioner motor 212 to move the board in the direction of the arrow 224 and apply its end edge to the plate 208.

504 547

Conveyor 200 moves the board to the justersaw which has had its adjustment 204 adjusted so that the saws divide the board longitudinally as required.

While individual conveyor 200 has been shown, it is obvious that two separate conveyors can be used. One for carrying the board to the positioning station, including the rollers 216 and the plate 208 and a second conveyor which, depending on the function, is used to transport the boards through the jigsaws 202.

The invention has been described with the stem blocks fed substantially perpendicular to its longitudinal axis, but it is obvious that with necessary modifications, the stem blocks can be fed through the sensing equipment with the longitudinal axis substantially parallel to the feeding direction.

Thus, once the invention has been described, modifications will be apparent to those skilled in the art without departing from the spirit and spirit of the invention as defined in the appended claims.

504

547