WO2011123057A1 - Apparatus and methods for acquiring and analysing images of a pallet load - Google Patents

Apparatus and methods for acquiring and analysing images of a pallet load Download PDF

Info

Publication number
WO2011123057A1
WO2011123057A1 PCT/SG2010/000125 SG2010000125W WO2011123057A1 WO 2011123057 A1 WO2011123057 A1 WO 2011123057A1 SG 2010000125 W SG2010000125 W SG 2010000125W WO 2011123057 A1 WO2011123057 A1 WO 2011123057A1
Authority
WO
WIPO (PCT)
Prior art keywords
pallet load
image
predetermined pattern
projection
pallet
Prior art date
Application number
PCT/SG2010/000125
Other languages
French (fr)
Inventor
Dmitty Nechiporenko
Andrew Conley
Original Assignee
Azimuth Intellectual Products Pte Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Azimuth Intellectual Products Pte Ltd filed Critical Azimuth Intellectual Products Pte Ltd
Priority to PCT/SG2010/000125 priority Critical patent/WO2011123057A1/en
Publication of WO2011123057A1 publication Critical patent/WO2011123057A1/en

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01BMEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00Measuring arrangements characterised by the use of optical techniques
    • G01B11/24Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/50Depth or shape recovery
    • G06T7/521Depth or shape recovery from laser ranging, e.g. using interferometry; from the projection of structured light
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/10Image acquisition
    • G06V10/12Details of acquisition arrangements; Constructional details thereof
    • G06V10/14Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/145Illumination specially adapted for pattern recognition, e.g. using gratings
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30108Industrial image inspection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00Indexing scheme relating to image or video recognition or understanding
    • G06V2201/06Recognition of objects for industrial automation

Definitions

  • the invention relates to an apparatus and a method for acquiring an image of a pallet load.
  • the invention also relates to an apparatus and a method for deriving information relating to an arrangement of a pallet load.
  • the invention also extends to machine- (computer-) readable media having stored thereon machine-readable instructions for executing, in a machine, the aforementioned methods.
  • the invention has particular, but not exclusive, application in implementing
  • machine vision techniques for capturing images of a pallet of goods cartons or goods packages, and may be subjected to subsequent analysis.
  • Automated asset tracking for palleted goods and the like offers significant cost savings in terms of resources when compared to known techniques for manual checking and tracking of these types of goods.
  • the present applicant has proposed techniques for acquisition of images of pallet loads. Additionally, the present applicant has proposed techniques for the analysis of these images and the creation of data models of the goods packages and/or pallet loads.
  • commonly- owned International Patent Application No. PCT/SG 2009/000157 discloses techniques which relate to acquisition of an image of a pallet load at sufficient resolution to enable barcodes with pitch less than 1mm to be discerned clearly.
  • the images acquired can be of at least part of, or all of, the pallet load; for instance, the image may be of a goods package disposed upon the pallet or a part thereof.
  • the disclosed techniques facilitate all types of image analysis including optical character recognition, barcode recognition and regular and irregular shape detection (for example, damage detection).
  • a stack 900 of goods packages 902a, 902b and 902c is illustrated in elevational view.
  • the stack does not have a package - that is, thre is a gap in space 904 - and this may be difficult to detect from a two-dimensional image.
  • Three-dimensional laser scanner systems which can be used for surface mapping of an object. Such systems are used to create an exact geometric reconstruction of the shape of a surface - including minute surface distortions - and are considered to be "overkill” for the likes of say, detecting a missing goods package in the space 904 of Figure 9. As mentioned, these are capable of mapping minute distortions in a surface and are impractically expensive in the present context. The same can be said for laser rangefinders and ultrasonic distance sensors.
  • an image of the pallet with the predetermine pattern thereon can be used to derive useful information relating to the arrangement of the pallet load as will be described in detail below. For instance, information relating to the arrangement of the stacking of the packages including whether any packages are missing from the stack may be derived. Further, chaotic and irregular arrangement of goods packages may also be detected which, absent said detection, may otherwise lead to spurious results.
  • depth of field information for example, how far from the camera the goods packages are - can be utilised to scale data elements from the images acquired.
  • Examples of the types of data elements include barcodes, logos, text strings on the packaging and the like extracted from images of the pallet load and from this scaling, more reliable and accurate results may be realised.
  • the techniques disclosed herein provide acceptable results at a significantly reduced cost.
  • Figure 1 is a perspective diagram illustrating a first apparatus for acquiring an image of a pallet load
  • Figure 2 is a layout diagram illustrating projection of a predetermined pattern on to a pallet load using the apparatus of Figure 1;
  • Figure 3 is a layout diagram illustrating a projection device suitable for use in the apparatus of Figure 1;
  • Figure 4 is a layout diagram illustrating projection of a predetermined pattern using the projection device of Figure 3;
  • Figure 5 is a block diagram illustrating a second apparatus for acquiring an image of a pallet load
  • Figure 6 is an architecture block diagram illustrating an apparatus for deriving information relating to an arrangement of a pallet load
  • Figure 7 is a layout diagram illustrating an image acquired by the apparatus of Figure l, Figure 2 or Figure 5 for analysis using the apparatus of Figure 6;
  • Figure 8 is a layout diagram illustrating a predetermined pattern projected on to a pallet load of irregularly stacked goods packages
  • Figure 9 is a layout diagram illustrating a stack of goods package stacked in an irregular fashion.
  • FIG. 1 illustrates a first apparatus for acquiring an image of a pallet load.
  • the apparatus comprises a projection device for projecting a
  • apparatus 100 comprises a linescan camera 102 for acquiring an image of a pallet load 112 disposed upon a pallet 110.
  • Pallet load 112 may have been placed in position on pallet 110 with the assistance of, say, a fork lift truck (not shown).
  • a pallet load 112 comprises of a plurality of goods packages 112a stacked upon pallet 110 in an orderly manner. In this case, the goods packages 112a are stacked in an orderly 3 x 3 x 3 matrix.
  • Linescan camera 102 is arranged to move vertically in plane 116.
  • linescan camera 102 is mounted for linear movement on supports 106a, 106b which comprise of linear actuators, having suitable drivers as will be known to the skilled person, and these are arranged upon a support base 108, Linescan camera 102 is moved from top to bottom (and vice versa) in the direction of the arrows 116.
  • the field of view (optical plane) of linescan camera 102 is incident on pallet load 112, although the field of view is not illustrated in Figure 1 for the sake of clarity. So, while travelling, linescan camera 102 acquires an image of the pallet load one line at a time. The result is a complete high-resolution (for example, 160 Mpixel) image.
  • a suitable linescan camera is one which is set to grab about 6 to 10 lines per millimetre with a field of view of a line being 1 pixel thick and 8000 pixels wide. However, other, for example, higher resolutions are also contemplated.
  • Apparatus 100 also comprises a projection device 104 for projecting a
  • Apparatus 100 also comprises a primary light source 103 for illuminating the pallet load 112.
  • Camera 102 may be provided with a polarising filter 101 for polarising light reflections from the pallet load 112.
  • projection device 104 projects the predetermined pattern on to the pallet load 112 at a projection angle.
  • Projection angle is an angle ⁇ which is a non-zero angle with respect to a normal N of a plane 114 of pallet load 112.
  • Plane 114 is a reference plane in three-dimensional space which can be considered to represent the "ideal" alignment of the pallet load 112 so that all surfaces of the goods packages 112a in the pallet load 112 facing apparatus 100 are substantially aligned in the vertical plane to represent a flat or substantially flat surface facing apparatus 100.
  • the linescan camera is arranged for linear movement in the vertical plane 116 but other arrangements, for example, other directions of movement, are not excluded.
  • the linescan camera 102 could be arranged for linear movement in the horizontal plane as an alternative or in addition to movement in the horizontal plane.
  • Figure 1 illustrates a linescan camera 102
  • other types of camera may be used where appropriate.
  • Other suitable types of camera include a CCD matrix camera which can be stationary or, as discussed with reference to Figure 5, can also be moveable.
  • apparatus 100 also comprises a primary light source 103 for illuminating the pallet load 112 in the example of Figure 1.
  • Apparatus 100 is arranged for the primary light source 103 to be inactive when the projection device 104 is active when the apparatus 100 is acquiring the image of pallet load 112 with the predetermined pattern projected thereon, primary light source 103 is selectively deactivated so as not to obscure the projection of the predetermined pattern on pallet load 112.
  • the apparatus 100 is arranged for the linescan camera 102 to acquire a first image of the pallet load in a first scan when the primary light source 103 is active and for the linescan camera to acquire a second image of the pallet load 112 in a second scan when primary light source 103 is inactive and the projection device 104 is active.
  • primary light source is activated (i.e. switched on) during a first scanning motion of linescan camera (as described above) to acquire an image of pallet load 112 and all the information thereon (i.e. on the goods packages, such as barcodes, logos, labels and the like) for analysis of the image and/or data modelling thereof.
  • Projection device 104 may be either active or inactive (i.e. on or off) during the scan but given that the intensity of the illumination of the projection device 102 is relatively limited in comparison to the primary light source 103, the predetermined pattern being projected on to the side of the pallet load 112 will typically do little to interfere with the later recognition of this information in the acquired image(s).
  • deactivating of the projection device during the operation of scanning pallet load 112 by camera 102 may very well provide improved results.
  • the primary light source 103 is deactivated (i.e.
  • the projection device 104 is active so that when an image is acquired by camera 102 during the scan, the image contains image data relating to the predetermined pattern having been projected on to pallet load 112 which is readily discernible without being obfuscated by light from the primary light source 103 and the image may be processed in accordance with the techniques discussed with reference to Figures 6 and 7 below.
  • Figure 2a illustrates a pallet load 112 in elevational view disposed upon pallet 110.
  • a predetermined pattern 118 has been projected on side 117 of pallet load 112. It will be appreciated that pallet load side 117 comprises the collective sides of the individual goods packages 112a (not shown in Figure 2a for the purposes of clarity).
  • predetermined pattern 118 comprises a rectangle but other shapes including regular shapes such as circles, triangles and other polygons may also be used. It has been found that predetermined pattern shapes comprising at least in part of straight lines such as those found in regular shapes are particularly suitable to the present application. Indeed, this is discussed further in relation to Figure 3.
  • each of the sides of pre-determined pattern 118 have been projected onto side 117 of pallet load 112 using one or more linelasers.
  • Figure 2a shows pallet load 112 comprised of individual goods packages 112a in plan view disposed upon pallet 110. Also illustrated disposed between pallet 110 and camera 104 is plane 114 of the pallet load as described above with reference to Figure 1. As illustrated, plane 114 has a normal N in three-dimensional space and projection device 104 projects predetermined pattern 118 on side 117 of pallet load 112 with a field of projection 120. In the example Figure 2, projection field 120 represents the projection of the vertical lines of predetermined pattern 118 visible in Figure 2a using the one or more linelasers. As illustrated in Figure 2b, these lines are projected at an angle ⁇ to normal N of plane 114, the benefits of which will be apparent from the following discussion particularly in relation to Figures 6 and 7.
  • the projection lines 120 are oriented at an angle ⁇ from the normal N in the vertical plane 119.
  • the horizontal lines of predetermined pattern 118 visible in Figure 2a are also projected at an angle displaced from the normal N of plane 114 in the horizontal plane (not shown in Figure 2). This angle of projection can be at ⁇ or at another angle.
  • the projection angle ⁇ is around one degree from the normal N, but other suitable angles may also be used.
  • the projection device 104 comprises a plurality of linelasers 308, 310 arranged to project a laser line grid on the pallet load as the predetermined pattern 118, and one or more of the linelasers 308, 310 are arranged to project a laserline at the projection angle ⁇ .
  • device 102 comprises a plurality of linelasers mounted on a frame consisting of vertical and horizontal members 302, 304.
  • An array 306 of linelasers 308 is arranged in the horizontal axis adjacent (or supported on) horizontal member 304.
  • An array 309 of linelasers 310 is disposed in the vertical axis either fixed on or supported by vertical member 302.
  • One or more of linelasers 308, 310 are mounted for projection at an angle equal to or greater than ⁇ displaced from the normal N of plane 114 in the horizontal and vertical axes respectively.
  • the projection device 104 It is generally desirable for the projection device 104 to project the predetermined pattern at a relatively low intensity for reasons of cost-efficiency and, in instances where the projection device 104 is active during a scan of a linescan camera to acquire a principal image of the pallet load, to provide less interference in the main image so as not to obscure the data elements - logos, barcodes, etc. - with the predetermined pattern.
  • the projection device utilises linelasers such as in Figure 3
  • an intensity in the region of 5mW has been found to provide acceptable results. Parasitic light elements from the projection device in the main image are therefore obviated.
  • a fibre grating is positioned between a laser and the pallet load.
  • a laser will have a power in the region of lOOmW.
  • a predetermined pattern of dots or other shapes, depending on the nature of the fibre grating
  • the dots provide a regular pattern, e.g. the spacing of dots at certain distance from the grating is equal.
  • Such implementations provide for a relatively simple installations, with costing approximately the same, or perhaps even less, than with a grid of lasers.
  • Figure 4 shows in outline (for the purposes of clarity) pallet 110 having pallet load 112 disposed thereon.
  • a laser grid 402 projected from device 102 comprises of vertical laserlines 404 and horizontal laserlines 406 projected on to side 117 of pallet load 112.
  • the image illustrated in Figure 4 is typical of an image captured when primary light source 103 is deactivated.
  • the edges of the goods packages 112a and other elements thereon are not captured in the image, or at least not in sufficient detail for to interfere with viewing of the predetermined patter 118.
  • the goods packages of pallet load 112 have been arranged on pallet 110 in an orderly fashion so laser grid 402 is represented as a regular grid with no discernible distortions therein.
  • Figure 5 illustrates a second apparatus for acquiring an image of a pallet load.
  • camera 102 (not shown) is mounted on a fixed support for acquisition of images of pallet load 112 disposed on pallet 110.
  • Projection device 104 is mounted on a support 500 moveable between a first position 502 and a second position 504 as indicated by arrow 503.
  • the apparatus is arranged for camera 102 to acquire a first image of pallet load 112 with projection device 104 at the first position 502, the first image being an image of the pallet load 112 with a first projection 506 of the pre-determined pattern projected thereon and to acquire a second image at the second position 504, the second image being an image of the pallet load 112 with a second projection 508 of the predetermined pattern projected thereon.
  • the predetermined pattern is projected in a field of projection 520a at the first position onto surface 117 of pallet load 112 and projected in field of projection 520b at the second position onto surface 117 of pallet load 112.
  • the field of projection has an angle of 2 ⁇ , corresponding to an angle of ⁇ on either side of the normal N.
  • projection device 104 is set up to project a predetermined pattern such as pattern 118 of Figure 2.
  • Projection fields 520a, 520b represent the projection of the vertical lines of predetermined pattern 118 visible in Figure 2a using the one or more linelasers. These lines are projected at an angle ⁇ (not shown) to normal N of plane 114.
  • the projection lines 520a, 520b are oriented at an angle ⁇ from the normal N in the vertical plane.
  • the horizontal lines of predetermined pattern 118 visible in Figure 2a are also projected at an angle displaced from the normal N of plane 114 in the horizontal plane (not shown in Figure 5). This angle of projection can be at ⁇ or at another angle.
  • the vertical (and the horizontal lines) need not all be projected at the same angle ⁇ .
  • one of the vertical or horizontal lines may be projected at a first angle ⁇ 1 and another of the vertical or horizontal lines may be projected at a second angle ⁇ 2.
  • the first projection is projected onto surface 117 of pallet load 112 as denoted by reference numeral 506.
  • the second projection is projected onto surface 117 of pallet load 112 as denoted by reference numeral 508.
  • the first projection and the second project may overlap, at least partially, in the area 509 of pallet load 112. As illustrated, the overlap area 509 is a partial overlap of projections 506, 508.
  • the overlap may be complete or at least substantially complete; that means that the area denoted 506 and the area denoted 508 are superimposed at at least substantially superimposed and the fields of projection 520a, 520 are coincident or at least substantially coincident on pallet load 112.
  • a benefit arising from such an arrangement is that a projection device 104 of a lower specification may be provided than would otherwise be needed to project a single projection of predetermined pattern over a large part of the pallet load, such as the combined area of areas 506 and 508. In moving from the first position 502 to the second position 504, this varies the projection of the predetermined pattern on to pallet load 112 so that in the two images, the pattern is projected on to different positions of pallet load 112.
  • projection device 104 is comprised of the device 104 illustrated in Figure 3, it will be appreciated that projection device 104 might then use fewer numbers of lasers to project a less dense pattern (e.g. a grid) on pallet 112 in each projection.
  • a relatively simple mechanical device to enable movement of support 500 from first position 502 to second position 504, two images (i.e. images of parts 506 and 508 of pallet load 112) are acquired which can be joined together using known techniques such as image stitching, superimposing of the images or other suitable image processing techniques.
  • a denser grid of laser projection lines is then obtained in the resultant combined image for a lower number of lasers. Broader coverage - i.e. across a greater area of surface 117 - of the predetermined pattern on the pallet load 112 may also be obtained.
  • Each of the first and second projections of the predetermined pattern might be restricted to the first and second parts 506, 508 of the pallet load but they may overlap, or even be over a broader range such as over the whole side 117 (or substantially the whole side) of pallet load 112.
  • projection device 104 also utilises linelasers but it will be appreciated other types of projection devices may also be used to project other patterns and images on to pallet load 112.
  • Figure 5 is given in the context of camera 102 being mounted upon a fixed support, it will be appreciated that this is not essential.
  • the camera 102 being a linescan camera arranged for planar movement to acquire (or "scan") an image of pallet load 112 or the camera being positioned on the (or another) support 500.
  • the apparatus may be configured for the camera to acquire a first image of the pallet load at the first position and a second image of the pallet load at the second position.
  • the images can be of part or all of the pallet load, and joined together, as discussed above.
  • Apparatus 600 principally comprises a processor (such as a microprocessor) 602 and a memory 604 for storing instructions such as computer code thereon.
  • the apparatus is configured under control of processor 502 to execute the instructions 606 stored in memory 604 to analyse an image of the pallet load 112 having a predetermined pattern 118 projected thereon. It is to be remembered that the predetermined pattern 118 has been projected on to the pallet load 112 at a projection angle ⁇ , the projection angle being a non-zero angle relative to a normal N of a plane 114 of pallet load 112.
  • Apparatus 600 is also configured to recognise a pattern component of the predetermined pattern 118 in the image and to derive information relating to the arrangement of the pallet load 118 from the pattern component.
  • the image is stored in memory 612 as received from an external device such as apparatus 100/camera 102 of Figure 1 through input/output module 610.
  • frame grabber 614 is configured to receive a direct image feed from camera 102 of Figure 1 to produce the image to be stored in memory 612.
  • the recognition of the pattern component is performed by module 608, a suitable type of which is an edge detection module which can be implemented to recognise edges in the predetermined pattern as projected on to the side of pallet load 112.
  • image combination module 616 operates to combine the images acquired by the apparatus of Figure 5 (or similar).
  • apparatus 600 is configured to form the image as a combined image, the combined image being a combination of the first image of the first part 506 of pallet load 112 and the second image of the second part 508 of pallet load 112. Further, apparatus 600 is configured to combine the first image comprising a first projection of the predetermined pattern 118 with the second image comprising a second projection of the predetermined pattern 118.
  • Suitable types of image combination techniques are discussed above, and include the likes of image stitching.
  • Figure 7a illustrates an elevational view of a pallet load 112 disposed on a pallet 110.
  • Predetermined pattern 118 has been projected on to side 117 of pallet load 112.
  • a gap 704 appears where no goods package 112a has been positioned. This is similar to the discussion given above in relation to gap 904 of Figure 9.
  • Such gaps cause an irregularity/distortion of predetermined pattern 118 such that an image component 118b is produced.
  • apparatus 600 is configured to perform a comparison of the pattern component 118b with a reference pattern component and to derive the information relating to the arrangement of the pallet load from the comparison.
  • the image component 118b is compared with a reference image component 118c (shown in a chain line).
  • the reference component 118c represents what apparatus 600 would expect for the remainder of predetermined pattern 118 if it had been projected upon a regular surface 117 without there being a gap 704; that is, if another goods package 112a was disposed in the gap 704. Additionally or alternatively, the reference pattern component might be the undistorted portion 118 from which the calculations may be made.
  • apparatus 600 is configured to determine a distance from a reference point of an object in the pallet load. So, the distance of the package behind the gap upon which image component 118b has been projected can be determined from a reference point such as the camera 102 or the projection device 104.
  • apparatus 600 determines a first object (a goods package on which image component 118 is projected) in the pallet load is at a first distance from the reference point and to determine a second object (a goods package on which image component 118b is projected) in the pallet load is at a second distance from the reference point. Therefore "gaps" 704 in the pallet load can be detected.
  • apparatus 600 is able to detect irregularities in the predetermined pattern to detect distance information and/or chaotic stacking. In calculating the distance information, if there is sufficient distortion/irregularity above a threshold, apparatus 600 can detect that a package is missing and/or stacked in an irregular fashion.
  • apparatus 600 is configured to determine objects in the pallet load are disposed in an irregular arrangement, such as being stacked chaotically, or with gaps, such as the gap 704 in Figure 7.
  • An irregular arrangement is best understood with respect to Figure 8 which shows a laser grid 800 which has been projected on to a pallet load (not shown) disposed upon a pallet (also not shown) in an irregular fashion.
  • Grid 800 comprises image component 804 where a "square" of grid appears larger than other squares in the grid. Given that the predetermined pattern fans out form the projection device 104, this provides an indicator that the surface on which this image component 804 has been projected is at a distance farther from the reference point - for example, the camera or the projection device itself - when compared with the adjacent linelasers.
  • Image component 806 is rather fragmented indicating an irregular stacking of goods packages in this area. For example, one of the goods packages may have been twisted or rotated such that its surface is not properly aligned with the other surfaces of the other goods packages 112a in the pallet load 112.
  • Image component 810 comprises a square of grid which appears smaller than other squares in grid 800. As noted above, given that the
  • this provides an indicator that the surface on which this component is projected is at a distance nearer the reference point than the surrounding areas where the other adjacent linelaser portions are projected.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Multimedia (AREA)
  • Optics & Photonics (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

An apparatus for acquiring an image of a pallet load comprises a projection device for projecting a predetermined pattern on to the pallet load at a projection angle. The projection angle is a non-zero angle relative to a normal of a plane of the pallet load. The apparatus also comprises a camera for acquiring an image of the pallet load having the predetermined pattern projected thereon. An apparatus for deriving information relating to an arrangement of a pallet load is configured to analyse an image of the pallet load having a predetermined pattern projected thereon at a non- zero projection angle. The apparatus recognises a pattern component of the predetermined pattern in the image and derives information relating to the arrangement of the pallet load from the pattern component.

Description

APPARATUS AND METHODS FOR ACQUIRING AND ANALYSING IMAGES OF A
PALLET LOAD
The invention relates to an apparatus and a method for acquiring an image of a pallet load. The invention also relates to an apparatus and a method for deriving information relating to an arrangement of a pallet load. The invention also extends to machine- (computer-) readable media having stored thereon machine-readable instructions for executing, in a machine, the aforementioned methods.
The invention has particular, but not exclusive, application in implementing
"machine vision" techniques for capturing images of a pallet of goods cartons or goods packages, and may be subjected to subsequent analysis.
Automated asset tracking for palleted goods and the like offers significant cost savings in terms of resources when compared to known techniques for manual checking and tracking of these types of goods. The present applicant has proposed techniques for acquisition of images of pallet loads. Additionally, the present applicant has proposed techniques for the analysis of these images and the creation of data models of the goods packages and/or pallet loads. For instance, commonly- owned International Patent Application No. PCT/SG 2009/000157 discloses techniques which relate to acquisition of an image of a pallet load at sufficient resolution to enable barcodes with pitch less than 1mm to be discerned clearly. The images acquired can be of at least part of, or all of, the pallet load; for instance, the image may be of a goods package disposed upon the pallet or a part thereof. In addition, the disclosed techniques facilitate all types of image analysis including optical character recognition, barcode recognition and regular and irregular shape detection (for example, damage detection).
However, if goods packages are stacked on the pallet in an irregular fashion this can be difficult to detect leading to spurious and/or erroneous results in the analysis of the images and the possible subsequent data modelling. Referring to Figure 9, a stack 900 of goods packages 902a, 902b and 902c is illustrated in elevational view. In the space marked 904, the stack does not have a package - that is, thre is a gap in space 904 - and this may be difficult to detect from a two-dimensional image.
Should another package be positioned behind this space, certain image acquisition techniques may not detect this fact and erroneously ascribe any data extracted from the image to a goods package in the space 904, instead of to its true goods package which is actually located behind space 904.. As noted, this leads to misleading results. For instance, any subsequent data modelling (or human analysis of the images) might incorrectly consider the box was positioned in the space 904 thus leading to inventory control inaccuracies.
It is also sometimes difficult to detect how close a goods package is to a camera acquiring an image of the goods package/pallet load and this is particularly true when the goods packages are stacked in an irregular or disorderly fashion. Any machine vision techniques may return false results due to characters being mistaken for smaller or larger than they really are without having any depth of field
information in the analysis of the images acquired. For instance, in the foregoing example relating to the space 904 in the stack 900 of Figure 9, if another goods package is behind the space in 904 an image acquisition apparatus such as a camera will read information from the package but given that the other goods package is farther away from the camera data elements extracted from the image relating to that package will be smaller and/or harder to read. Alternatively, if characters (or other image components) are present on a goods package arranged at an angle to the camera acquiring the image of the goods package, the characters may be difficult to recognise. Such issues may lead to incorrect analysis if the results obtained from the image are not in accordance with expected dimensional information such as font size, package label sizes and the like.
Yet a further problem arises when cartons or packages in a stack of goods packages are damaged where holes do not necessarily appear in the packages but simply there may be geometrical distortions causing further difficulties in reading and analysing information/data on the packages themselves.
Three-dimensional laser scanner systems, which can be used for surface mapping of an object. Such systems are used to create an exact geometric reconstruction of the shape of a surface - including minute surface distortions - and are considered to be "overkill" for the likes of say, detecting a missing goods package in the space 904 of Figure 9. As mentioned, these are capable of mapping minute distortions in a surface and are impractically expensive in the present context. The same can be said for laser rangefinders and ultrasonic distance sensors.
The invention is defined in the independent claims. Some optional features of the invention are defined in the dependent claims.
When a predetermined pattern is projected on to the pallet load at a projection angle which is non-zero relative to a normal of a plane of the pallet load (as will be described further below), an image of the pallet with the predetermine pattern thereon can be used to derive useful information relating to the arrangement of the pallet load as will be described in detail below. For instance, information relating to the arrangement of the stacking of the packages including whether any packages are missing from the stack may be derived. Further, chaotic and irregular arrangement of goods packages may also be detected which, absent said detection, may otherwise lead to spurious results.
Additionally, depth of field information - for example, how far from the camera the goods packages are - can be utilised to scale data elements from the images acquired. Examples of the types of data elements include barcodes, logos, text strings on the packaging and the like extracted from images of the pallet load and from this scaling, more reliable and accurate results may be realised. In comparison to, say, three-dimensional laser scanner systems, the techniques disclosed herein provide acceptable results at a significantly reduced cost.
The invention will now be described by way of example only and with reference to the accompanying drawings in which:
Figure 1 is a perspective diagram illustrating a first apparatus for acquiring an image of a pallet load;
Figure 2 is a layout diagram illustrating projection of a predetermined pattern on to a pallet load using the apparatus of Figure 1;
Figure 3 is a layout diagram illustrating a projection device suitable for use in the apparatus of Figure 1;
Figure 4 is a layout diagram illustrating projection of a predetermined pattern using the projection device of Figure 3;
Figure 5 is a block diagram illustrating a second apparatus for acquiring an image of a pallet load;
Figure 6 is an architecture block diagram illustrating an apparatus for deriving information relating to an arrangement of a pallet load;
Figure 7 is a layout diagram illustrating an image acquired by the apparatus of Figure l,Figure 2 or Figure 5 for analysis using the apparatus of Figure 6;
Figure 8 is a layout diagram illustrating a predetermined pattern projected on to a pallet load of irregularly stacked goods packages;
Figure 9 is a layout diagram illustrating a stack of goods package stacked in an irregular fashion.
Turning first to Figure 1, this illustrates a first apparatus for acquiring an image of a pallet load. The apparatus comprises a projection device for projecting a
predetermined pattern on to the pallet load at a projection angle, the projection angle being a non-zero angle relative to a normal of a plane of the pallet load. The apparatus also comprises a camera for acquiring an image of the pallet load having the predetermined pattern projected thereon. In the example of Figure 1, apparatus 100 comprises a linescan camera 102 for acquiring an image of a pallet load 112 disposed upon a pallet 110. Pallet load 112 may have been placed in position on pallet 110 with the assistance of, say, a fork lift truck (not shown). In the example of Figure 1, a pallet load 112 comprises of a plurality of goods packages 112a stacked upon pallet 110 in an orderly manner. In this case, the goods packages 112a are stacked in an orderly 3 x 3 x 3 matrix.
Linescan camera 102 is arranged to move vertically in plane 116. In the example, linescan camera 102 is mounted for linear movement on supports 106a, 106b which comprise of linear actuators, having suitable drivers as will be known to the skilled person, and these are arranged upon a support base 108, Linescan camera 102 is moved from top to bottom (and vice versa) in the direction of the arrows 116. The field of view (optical plane) of linescan camera 102 is incident on pallet load 112, although the field of view is not illustrated in Figure 1 for the sake of clarity. So, while travelling, linescan camera 102 acquires an image of the pallet load one line at a time. The result is a complete high-resolution (for example, 160 Mpixel) image. A suitable linescan camera is one which is set to grab about 6 to 10 lines per millimetre with a field of view of a line being 1 pixel thick and 8000 pixels wide. However, other, for example, higher resolutions are also contemplated.
Apparatus 100 also comprises a projection device 104 for projecting a
predetermined pattern on to the pallet load at the projection angle. This will be discussed in more detail with reference to Figure 2. As such, the field of projection of projection device 104 is also omitted for the sake of clarity from Figure 1.
Apparatus 100 also comprises a primary light source 103 for illuminating the pallet load 112. Camera 102 may be provided with a polarising filter 101 for polarising light reflections from the pallet load 112. As noted, projection device 104 projects the predetermined pattern on to the pallet load 112 at a projection angle. Projection angle is an angle Θ which is a non-zero angle with respect to a normal N of a plane 114 of pallet load 112. Plane 114 is a reference plane in three-dimensional space which can be considered to represent the "ideal" alignment of the pallet load 112 so that all surfaces of the goods packages 112a in the pallet load 112 facing apparatus 100 are substantially aligned in the vertical plane to represent a flat or substantially flat surface facing apparatus 100.
It will be appreciated that in the example of Figure 1, the linescan camera is arranged for linear movement in the vertical plane 116 but other arrangements, for example, other directions of movement, are not excluded. For example, the linescan camera 102 could be arranged for linear movement in the horizontal plane as an alternative or in addition to movement in the horizontal plane.
Although the example of Figure 1 illustrates a linescan camera 102, it will also be appreciated that other types of camera may be used where appropriate. Other suitable types of camera include a CCD matrix camera which can be stationary or, as discussed with reference to Figure 5, can also be moveable.
As noted, apparatus 100 also comprises a primary light source 103 for illuminating the pallet load 112 in the example of Figure 1. Apparatus 100 is arranged for the primary light source 103 to be inactive when the projection device 104 is active when the apparatus 100 is acquiring the image of pallet load 112 with the predetermined pattern projected thereon, primary light source 103 is selectively deactivated so as not to obscure the projection of the predetermined pattern on pallet load 112. In the scanning operation using a linescan camera, the apparatus 100 is arranged for the linescan camera 102 to acquire a first image of the pallet load in a first scan when the primary light source 103 is active and for the linescan camera to acquire a second image of the pallet load 112 in a second scan when primary light source 103 is inactive and the projection device 104 is active. In the example of Figure 1 primary light source is activated (i.e. switched on) during a first scanning motion of linescan camera (as described above) to acquire an image of pallet load 112 and all the information thereon (i.e. on the goods packages, such as barcodes, logos, labels and the like) for analysis of the image and/or data modelling thereof. Projection device 104 may be either active or inactive (i.e. on or off) during the scan but given that the intensity of the illumination of the projection device 102 is relatively limited in comparison to the primary light source 103, the predetermined pattern being projected on to the side of the pallet load 112 will typically do little to interfere with the later recognition of this information in the acquired image(s). Of course, deactivating of the projection device during the operation of scanning pallet load 112 by camera 102 may very well provide improved results. In the second scanning action, the primary light source 103 is deactivated (i.e. switched off) while the projection device 104 is active so that when an image is acquired by camera 102 during the scan, the image contains image data relating to the predetermined pattern having been projected on to pallet load 112 which is readily discernible without being obfuscated by light from the primary light source 103 and the image may be processed in accordance with the techniques discussed with reference to Figures 6 and 7 below.
Figure 2a illustrates a pallet load 112 in elevational view disposed upon pallet 110. A predetermined pattern 118 has been projected on side 117 of pallet load 112. It will be appreciated that pallet load side 117 comprises the collective sides of the individual goods packages 112a (not shown in Figure 2a for the purposes of clarity). In the example of Figure 2a, predetermined pattern 118 comprises a rectangle but other shapes including regular shapes such as circles, triangles and other polygons may also be used. It has been found that predetermined pattern shapes comprising at least in part of straight lines such as those found in regular shapes are particularly suitable to the present application. Indeed, this is discussed further in relation to Figure 3. However, it will be appreciated that practically any predetermined pattern can be used including images, but use of regular shapes with straight lines may provide results which are less complex to interpret. In the view of Figure 2a, the individual edge lines of the goods packages 112a of pallet load 112 have been omitted for the sake of clarity. Thus, the view in Figure 2a might be considered to represent the image acquired by the linescan camera 102 in a scan when primary light source 103 has been deactivated and projection device 104 has been activated, as discussed above with reference to Figure 1. Thus the advantage of deactivating primary light source 103 during the scan is readily apparent in that camera 102 acquires an image showing principally predetermined pattern 118 and simplifying the processing techniques required in order to analyse the predetermined pattern 118. Indeed, the outline of pallet load 112 (shown in Figure 2a by reference numeral 112b) may not be visible at all and outline 112b is shown in the view of Figure 2A for the purposes of assisting the reader's understanding.
In the example of Figure 2, each of the sides of pre-determined pattern 118 have been projected onto side 117 of pallet load 112 using one or more linelasers.
Figure 2a shows pallet load 112 comprised of individual goods packages 112a in plan view disposed upon pallet 110. Also illustrated disposed between pallet 110 and camera 104 is plane 114 of the pallet load as described above with reference to Figure 1. As illustrated, plane 114 has a normal N in three-dimensional space and projection device 104 projects predetermined pattern 118 on side 117 of pallet load 112 with a field of projection 120. In the example Figure 2, projection field 120 represents the projection of the vertical lines of predetermined pattern 118 visible in Figure 2a using the one or more linelasers. As illustrated in Figure 2b, these lines are projected at an angle Θ to normal N of plane 114, the benefits of which will be apparent from the following discussion particularly in relation to Figures 6 and 7. In the view of Figure 2b, the projection lines 120 are oriented at an angle Θ from the normal N in the vertical plane 119. However, it will be appreciated that the horizontal lines of predetermined pattern 118 visible in Figure 2a are also projected at an angle displaced from the normal N of plane 114 in the horizontal plane (not shown in Figure 2). This angle of projection can be at Θ or at another angle.
In the example of Figure 2, the projection angle Θ is around one degree from the normal N, but other suitable angles may also be used. Turning to Figure 3, a projection device 104 suitable for use with the apparatus of 100 of Figure 1 and Figure 2 is illustrated. The projection device 104 comprises a plurality of linelasers 308, 310 arranged to project a laser line grid on the pallet load as the predetermined pattern 118, and one or more of the linelasers 308, 310 are arranged to project a laserline at the projection angle Θ. Thus, device 102 comprises a plurality of linelasers mounted on a frame consisting of vertical and horizontal members 302, 304. An array 306 of linelasers 308 is arranged in the horizontal axis adjacent (or supported on) horizontal member 304. An array 309 of linelasers 310 is disposed in the vertical axis either fixed on or supported by vertical member 302. One or more of linelasers 308, 310 (for example, all of the linelasers) are mounted for projection at an angle equal to or greater than Θ displaced from the normal N of plane 114 in the horizontal and vertical axes respectively.
It is generally desirable for the projection device 104 to project the predetermined pattern at a relatively low intensity for reasons of cost-efficiency and, in instances where the projection device 104 is active during a scan of a linescan camera to acquire a principal image of the pallet load, to provide less interference in the main image so as not to obscure the data elements - logos, barcodes, etc. - with the predetermined pattern. Where the projection device utilises linelasers such as in Figure 3, an intensity in the region of 5mW has been found to provide acceptable results. Parasitic light elements from the projection device in the main image are therefore obviated.
Use of non-Gaussian linelasers gives a more even spread of illumination across the laser beam width, rather than lasers with a Guassian distribution beam field which have higher intensity in the centre of the beam field.
One other way to generate the pre-determined pattern for projection is a laser and a diffraction grating to generate a pattern of lasers on the pallet load. Of course, such techniques can also be implemented with multiple lasers and/or multiple diffraction gratings. In such implementations, a fibre grating is positioned between a laser and the pallet load. Typically, such a laser will have a power in the region of lOOmW. When the laser is activated, a predetermined pattern of dots (or other shapes, depending on the nature of the fibre grating) will appear on pallet. The dots provide a regular pattern, e.g. the spacing of dots at certain distance from the grating is equal. When the dots fan out from the fibre grating, the distance between dots projected on a surface at a farther distance from the grating will be greater than a distance nearer the grating.
Such implementations provide for a relatively simple installations, with costing approximately the same, or perhaps even less, than with a grid of lasers.
Figure 4 shows in outline (for the purposes of clarity) pallet 110 having pallet load 112 disposed thereon. A laser grid 402 projected from device 102 comprises of vertical laserlines 404 and horizontal laserlines 406 projected on to side 117 of pallet load 112. The image illustrated in Figure 4 is typical of an image captured when primary light source 103 is deactivated. Thus, the edges of the goods packages 112a and other elements thereon (labels, logos, etc.) are not captured in the image, or at least not in sufficient detail for to interfere with viewing of the predetermined patter 118. In the example, the goods packages of pallet load 112 have been arranged on pallet 110 in an orderly fashion so laser grid 402 is represented as a regular grid with no discernible distortions therein.
Figure 5 illustrates a second apparatus for acquiring an image of a pallet load. In the example of Figure 5, camera 102 (not shown) is mounted on a fixed support for acquisition of images of pallet load 112 disposed on pallet 110. Projection device 104 is mounted on a support 500 moveable between a first position 502 and a second position 504 as indicated by arrow 503. The apparatus is arranged for camera 102 to acquire a first image of pallet load 112 with projection device 104 at the first position 502, the first image being an image of the pallet load 112 with a first projection 506 of the pre-determined pattern projected thereon and to acquire a second image at the second position 504, the second image being an image of the pallet load 112 with a second projection 508 of the predetermined pattern projected thereon. The predetermined pattern is projected in a field of projection 520a at the first position onto surface 117 of pallet load 112 and projected in field of projection 520b at the second position onto surface 117 of pallet load 112. In each position, the field of projection has an angle of 2Θ, corresponding to an angle of Θ on either side of the normal N. In the example of Figure 5, projection device 104 is set up to project a predetermined pattern such as pattern 118 of Figure 2. Projection fields 520a, 520b represent the projection of the vertical lines of predetermined pattern 118 visible in Figure 2a using the one or more linelasers. These lines are projected at an angle Θ (not shown) to normal N of plane 114. In the view of Figure 5, the projection lines 520a, 520b are oriented at an angle Θ from the normal N in the vertical plane.
However, it will be appreciated that the horizontal lines of predetermined pattern 118 visible in Figure 2a are also projected at an angle displaced from the normal N of plane 114 in the horizontal plane (not shown in Figure 5). This angle of projection can be at Θ or at another angle.
It will also be appreciated that the vertical (and the horizontal lines) need not all be projected at the same angle Θ. For instance, one of the vertical or horizontal lines may be projected at a first angle Θ1 and another of the vertical or horizontal lines may be projected at a second angle Θ2.
The first projection is projected onto surface 117 of pallet load 112 as denoted by reference numeral 506. The second projection is projected onto surface 117 of pallet load 112 as denoted by reference numeral 508. The first projection and the second project may overlap, at least partially, in the area 509 of pallet load 112. As illustrated, the overlap area 509 is a partial overlap of projections 506, 508.
However, the overlap may be complete or at least substantially complete; that means that the area denoted 506 and the area denoted 508 are superimposed at at least substantially superimposed and the fields of projection 520a, 520 are coincident or at least substantially coincident on pallet load 112. A benefit arising from such an arrangement is that a projection device 104 of a lower specification may be provided than would otherwise be needed to project a single projection of predetermined pattern over a large part of the pallet load, such as the combined area of areas 506 and 508. In moving from the first position 502 to the second position 504, this varies the projection of the predetermined pattern on to pallet load 112 so that in the two images, the pattern is projected on to different positions of pallet load 112.
If, say, projection device 104 is comprised of the device 104 illustrated in Figure 3, it will be appreciated that projection device 104 might then use fewer numbers of lasers to project a less dense pattern (e.g. a grid) on pallet 112 in each projection. Thus, by introduction of a relatively simple mechanical device to enable movement of support 500 from first position 502 to second position 504, two images (i.e. images of parts 506 and 508 of pallet load 112) are acquired which can be joined together using known techniques such as image stitching, superimposing of the images or other suitable image processing techniques. A denser grid of laser projection lines is then obtained in the resultant combined image for a lower number of lasers. Broader coverage - i.e. across a greater area of surface 117 - of the predetermined pattern on the pallet load 112 may also be obtained.
Each of the first and second projections of the predetermined pattern might be restricted to the first and second parts 506, 508 of the pallet load but they may overlap, or even be over a broader range such as over the whole side 117 (or substantially the whole side) of pallet load 112.
It will be appreciated, of course, that the areas 506, 508 need not overlap in the region 509 as illustrated in Figure 5 and areas 506, 508 may be separate from one another. In Figure 5, projection device 104 also utilises linelasers but it will be appreciated other types of projection devices may also be used to project other patterns and images on to pallet load 112.
Although Figure 5 is given in the context of camera 102 being mounted upon a fixed support, it will be appreciated that this is not essential. For instance, various configurations for camera 102 are possible, and include the camera 102 being a linescan camera arranged for planar movement to acquire (or "scan") an image of pallet load 112 or the camera being positioned on the (or another) support 500. If the camera is positioned on the or another support, then the apparatus may be configured for the camera to acquire a first image of the pallet load at the first position and a second image of the pallet load at the second position. The images can be of part or all of the pallet load, and joined together, as discussed above.
Referring now to Figure 6, an apparatus 600 for deriving information relating to an arrangement of a pallet load 112 is illustrated. Apparatus 600 principally comprises a processor (such as a microprocessor) 602 and a memory 604 for storing instructions such as computer code thereon. The apparatus is configured under control of processor 502 to execute the instructions 606 stored in memory 604 to analyse an image of the pallet load 112 having a predetermined pattern 118 projected thereon. It is to be remembered that the predetermined pattern 118 has been projected on to the pallet load 112 at a projection angle Θ, the projection angle being a non-zero angle relative to a normal N of a plane 114 of pallet load 112. Apparatus 600 is also configured to recognise a pattern component of the predetermined pattern 118 in the image and to derive information relating to the arrangement of the pallet load 118 from the pattern component. In the example of Figure 6, the image is stored in memory 612 as received from an external device such as apparatus 100/camera 102 of Figure 1 through input/output module 610. In one implementation, frame grabber 614 is configured to receive a direct image feed from camera 102 of Figure 1 to produce the image to be stored in memory 612. The recognition of the pattern component is performed by module 608, a suitable type of which is an edge detection module which can be implemented to recognise edges in the predetermined pattern as projected on to the side of pallet load 112. Edge detection techniques are particularly useful when the predetermined pattern is a regular pattern composed of straight lines as would be produced by a linelaser. In one implementation, image combination module 616 operates to combine the images acquired by the apparatus of Figure 5 (or similar). In such implementations, apparatus 600 is configured to form the image as a combined image, the combined image being a combination of the first image of the first part 506 of pallet load 112 and the second image of the second part 508 of pallet load 112. Further, apparatus 600 is configured to combine the first image comprising a first projection of the predetermined pattern 118 with the second image comprising a second projection of the predetermined pattern 118. Suitable types of image combination techniques are discussed above, and include the likes of image stitching.
Turning now to Figure 7, an analysis of an image of pallet load 112 acquired using, say, the apparatus of Figure 1 is now described. Figure 7a illustrates an elevational view of a pallet load 112 disposed on a pallet 110. Predetermined pattern 118 has been projected on to side 117 of pallet load 112. However, in the pallet load 112 a gap 704 appears where no goods package 112a has been positioned. This is similar to the discussion given above in relation to gap 904 of Figure 9. Such gaps cause an irregularity/distortion of predetermined pattern 118 such that an image component 118b is produced. Referring to the detail of Figure 7b, because of the angle of projection Θ of the predetermined pattern and the difference in distance between the projection device of the surface upon which image component 118b has been projected when compared with the distance from the projection device image component 118 has been projected on the projected lines of the predetermined pattern "fan" out with distance from the projection device so that the image component 118b projected on the package behind the gap appears to be displaced from the main part of predetermined pattern 118. Accordingly, horizontal line of image component 118b is displaced from the horizontal line 118 by a distance dl where a break in the acquired image relating to the predetermined pattern occurs at the last edge 704a where the gap begins. Similarly, there is a displacement between predetermined pattern 118 and the vertical line component of image component 118b measured by distance d2. The amount of respective displacements dl and/or d2 allows apparatus 600 to derive distance information relating to the surface of the goods package upon which image component 118b is projected. Thus, apparatus 600 is configured to perform a comparison of the pattern component 118b with a reference pattern component and to derive the information relating to the arrangement of the pallet load from the comparison. In this case, the image component 118b is compared with a reference image component 118c (shown in a chain line). The reference component 118c represents what apparatus 600 would expect for the remainder of predetermined pattern 118 if it had been projected upon a regular surface 117 without there being a gap 704; that is, if another goods package 112a was disposed in the gap 704. Additionally or alternatively, the reference pattern component might be the undistorted portion 118 from which the calculations may be made. Thus, apparatus 600 is configured to determine a distance from a reference point of an object in the pallet load. So, the distance of the package behind the gap upon which image component 118b has been projected can be determined from a reference point such as the camera 102 or the projection device 104. That is, apparatus 600 determines a first object (a goods package on which image component 118 is projected) in the pallet load is at a first distance from the reference point and to determine a second object (a goods package on which image component 118b is projected) in the pallet load is at a second distance from the reference point. Therefore "gaps" 704 in the pallet load can be detected.
During calibration of the apparatuses 100, 500 and 600, a distance from the respective apparatus to the pallet load is measured and, from this information, distance information can be readily extrapolated. Alternatively, a reasonable approximation can be derived without this distance information from calibration, when the distance comparison is made. When implementing techniques without using a reference distance, apparatus 600 is able to detect irregularities in the predetermined pattern to detect distance information and/or chaotic stacking. In calculating the distance information, if there is sufficient distortion/irregularity above a threshold, apparatus 600 can detect that a package is missing and/or stacked in an irregular fashion.
Effectively, apparatus 600 is configured to determine objects in the pallet load are disposed in an irregular arrangement, such as being stacked chaotically, or with gaps, such as the gap 704 in Figure 7. An irregular arrangement is best understood with respect to Figure 8 which shows a laser grid 800 which has been projected on to a pallet load (not shown) disposed upon a pallet (also not shown) in an irregular fashion. Grid 800 comprises image component 804 where a "square" of grid appears larger than other squares in the grid. Given that the predetermined pattern fans out form the projection device 104, this provides an indicator that the surface on which this image component 804 has been projected is at a distance farther from the reference point - for example, the camera or the projection device itself - when compared with the adjacent linelasers. Image component 806 is rather fragmented indicating an irregular stacking of goods packages in this area. For example, one of the goods packages may have been twisted or rotated such that its surface is not properly aligned with the other surfaces of the other goods packages 112a in the pallet load 112. Image component 810 comprises a square of grid which appears smaller than other squares in grid 800. As noted above, given that the
predetermined pattern fans out at the angle of projection, this provides an indicator that the surface on which this component is projected is at a distance nearer the reference point than the surrounding areas where the other adjacent linelaser portions are projected.
It will be appreciated that the invention has been described by way of example only. Various modifications may be made to the techniques described herein without departing from the spirit and scope of the appended claims. The disclosed techniques comprise techniques which may be provided any stand-alone manner, or in combination with one of those. Therefore, features described with respect to one technique may also be used and presented in combination with another technique.

Claims

CLAIMS:
1. Apparatus for acquiring an image of a pallet load, the apparatus comprising: a projection device for projecting a predetermined pattern onto the pallet load at a projection angle, the projection angle being a non-zero angle relative to a normal of a plane of the pallet load; and
a camera for acquiring an image of the pallet load having the predetermined pattern projected thereon.
2. The apparatus of claim 1, wherein the projection device comprises a plurality of linelasers arranged to project a laser line grid on the pallet load as the
predetermined pattern, one or more of the linelasers being arranged to project a laser line at the projection angle.
3. The apparatus of claim 1 or claim 2 comprising a primary light source for illuminating the pallet load, the apparatus being arranged for the primary light source to be inactive when the projection device is active for the apparatus to acquire the image of the pallet load with the predetermined pattern projected thereon.
4. The apparatus of claim 3, wherein the camera comprises a linescan camera, and the apparatus is arranged for the linescan camera to acquire a first image of the pallet load in a first scan when the primary light source is active, and for the linescan camera to acquire a second image of the pallet load in a second scan when the primary light source is inactive and the projection device is active.
5. The apparatus of any preceding claim, wherein the projection device is mounted on a support movable between a first position and a second position, and to project a first projection of the predetermined pattern at the first position and to project a second projection of the predetermined pattern at the second position.
6. The apparatus of claim 5, wherein the apparatus is arranged for the first projection and the second projection to overlap at least partially on the pallet load.
7. The apparatus of any preceding claim further comprising a polarising filter for the camera.
8. Apparatus for deriving information relating to an arrangement of a pallet load, the apparatus comprising:
a processor; and
a memory; wherein
the apparatus is configured, under control of the processor, to execute instructions stored in the memory to:
analyse an image of the pallet load having a predetermined pattern projected thereon, the predetermined pattern having been projected onto the pallet load at a projection angle, the projection angle being a non-zero angle relative to a normal of a plane of the pallet load;
recognise a pattern component of the predetermined pattern in the image; and
derive information relating to the arrangement of the pallet load from the pattern component.
9. The apparatus of claim 8 configured to perform a comparison of the pattern component with a reference pattern component, and to derive the information relating to the arrangement of the pallet load from the comparison.
10. The apparatus of claim 8 or claim 9 configured to determine a distance, from a reference point, of a first object in the pallet load.
11. The apparatus of claim 10 configured to determine a first object in the pallet load is at a first distance from the reference point and to determine a second object in the pallet load is at a second distance from the reference point.
12. The apparatus of any of claims 8 to 11 configured to determine objects in the pallet load are disposed in an irregular arrangement.
13. The apparatus of any of claims 8 to 12 configured to form the image as a combined image, the combined image being a combination of a first image of the pallet load and a second image of the pallet load.
14. The apparatus of claim 13 configured to combine the first image comprising a first projection of the predetermined pattern with the second image comprising a second projection of the predetermined pattern.
15. A method for acquiring an image of a pallet load, the method comprising: operating a projection device to project a predetermined pattern onto the pallet load at a projection angle, the projection angle being a non-zero angle relative to a normal of a plane of the pallet load; and
operating a camera to acquire an image of the pallet load having the predetermined pattern projected thereon.
16. A method for deriving information relating to an arrangement of a pallet load, the method being implemented in an apparatus comprising a processor, and comprising controlling the processor for the apparatus to:
analyse an image of the pallet load having a predetermined pattern projected thereon, the predetermined pattern having been projected onto the pallet load at a projection angle, the projection angle being a non-zero angle relative to a normal of a plane of the pallet load; recognise a pattern component of the predetermined pattern in the image; and
derive information relating to the arrangement of the pallet load from the pattern component.
17. A machine-readable medium, having stored thereon machine-readable instructions for executing, in a machine, a method for deriving information relating to an arrangement of a pallet load, the method comprising controlling the processor for the machine to:
analyse an image of the pallet load having a predetermined pattern projected thereon, the predetermined pattern having been projected onto the pallet load at a projection angle, the projection angle being a non-zero angle relative to a normal of a plane of the pallet load;
recognise a pattern component of the predetermined pattern in the image; and
derive information relating to the arrangement of the pallet load from the pattern component.
PCT/SG2010/000125 2010-03-30 2010-03-30 Apparatus and methods for acquiring and analysing images of a pallet load WO2011123057A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/SG2010/000125 WO2011123057A1 (en) 2010-03-30 2010-03-30 Apparatus and methods for acquiring and analysing images of a pallet load

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/SG2010/000125 WO2011123057A1 (en) 2010-03-30 2010-03-30 Apparatus and methods for acquiring and analysing images of a pallet load

Publications (1)

Publication Number Publication Date
WO2011123057A1 true WO2011123057A1 (en) 2011-10-06

Family

ID=44712499

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/SG2010/000125 WO2011123057A1 (en) 2010-03-30 2010-03-30 Apparatus and methods for acquiring and analysing images of a pallet load

Country Status (1)

Country Link
WO (1) WO2011123057A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102016102251A1 (en) * 2016-02-10 2017-08-10 Ltw Intralogistics Gmbh analysis station
WO2018118294A1 (en) * 2016-12-20 2018-06-28 Symbol Technologies, Llc Methods, systems and apparatus for segmenting objects

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03234491A (en) * 1990-02-09 1991-10-18 Daifuku Co Ltd Automatic loading device using robot
JPH06241719A (en) * 1993-02-23 1994-09-02 Daifuku Co Ltd Article position detection device
JPH10118975A (en) * 1996-10-18 1998-05-12 Komatsu Ltd Handling position recognizing method and recognizing device
JP2003097918A (en) * 2001-09-27 2003-04-03 Kanebo Ltd Program for processing image using image processing device, or computer, and recording medium storing the same

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03234491A (en) * 1990-02-09 1991-10-18 Daifuku Co Ltd Automatic loading device using robot
JPH06241719A (en) * 1993-02-23 1994-09-02 Daifuku Co Ltd Article position detection device
JPH10118975A (en) * 1996-10-18 1998-05-12 Komatsu Ltd Handling position recognizing method and recognizing device
JP2003097918A (en) * 2001-09-27 2003-04-03 Kanebo Ltd Program for processing image using image processing device, or computer, and recording medium storing the same

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
PATENT ABSTRACTS OF JAPAN *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102016102251A1 (en) * 2016-02-10 2017-08-10 Ltw Intralogistics Gmbh analysis station
WO2018118294A1 (en) * 2016-12-20 2018-06-28 Symbol Technologies, Llc Methods, systems and apparatus for segmenting objects
US10354411B2 (en) 2016-12-20 2019-07-16 Symbol Technologies, Llc Methods, systems and apparatus for segmenting objects
GB2571458A (en) * 2016-12-20 2019-08-28 Symbol Technologies Llc Methods, systems and apparatus for segmenting objects
GB2571458B (en) * 2016-12-20 2021-12-15 Symbol Technologies Llc Methods, systems and apparatus for segmenting objects

Similar Documents

Publication Publication Date Title
US20120057022A1 (en) Apparatus and method for acquiring an image of a pallet load
US7878402B2 (en) Decoding distorted symbols
US10438036B1 (en) System and method for reading and decoding ID codes on a curved, sloped and/or annular object
US9519810B2 (en) Calibration and self-test in automated data reading systems
US10417769B2 (en) Automatic mode switching in a volume dimensioner
US20170150129A1 (en) Dimensioning Apparatus and Method
US11410405B2 (en) Imaging-based sensor calibration
JP4685971B2 (en) Inspection system and inspection method
GB2564794A (en) Image-stitching for dimensioning
JP5438475B2 (en) Gap step measurement device, gap step measurement method, and program thereof
WO2020159866A1 (en) Object dimensioning system and method
EP3353705B1 (en) Compact imaging module with range finder
JP6275622B2 (en) Method and scanner for detecting the position and three-dimensional shape of a plurality of products on a running surface in a non-contact manner
JP2015114292A (en) Workpiece position information identification apparatus and workpiece position information identification method
CN107907055B (en) Pattern projection module, three-dimensional information acquisition system, processing device and measuring method
JP2013108933A (en) Information terminal device
KR20140016341A (en) Apparatus for detecting end portion position of strip-like body, and method for detecting end portion position of strip-like body
JP4338405B2 (en) Size measuring system and size measuring device
US20130182294A1 (en) Image scanning apparatus and methods of using the same
WO2011123057A1 (en) Apparatus and methods for acquiring and analysing images of a pallet load
JP5891399B2 (en) Work detection system setting method and work detection system using the same
US10907954B2 (en) Methods and systems for measuring dimensions of a 2-D object
JP2024002525A (en) Bar code detection system, method for detecting bar code, and bar code detection program
WO2011087454A1 (en) Apparatus and methods for manipulating and acquiring images of a pallet load
CA2536411C (en) Multiple axis multipoint non-contact measurement system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10849110

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 30.01.2013)

122 Ep: pct application non-entry in european phase

Ref document number: 10849110

Country of ref document: EP

Kind code of ref document: A1