(19)
(11)EP 3 572 996 A1

(12)EUROPEAN PATENT APPLICATION

(43)Date of publication:
27.11.2019 Bulletin 2019/48

(21)Application number: 18382347.5

(22)Date of filing:  21.05.2018
(51)Int. Cl.: 
G06Q 10/08  (2012.01)
G06Q 50/28  (2012.01)
B65D 19/44  (2006.01)
(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR
Designated Extension States:
BA ME
Designated Validation States:
KH MA MD TN

(71)Applicant: Automatizacion y Sistemas de Inspeccion en Linea Global, S.L.
33207 Gijón (ES)

(72)Inventors:
  • Pérez Méndez, Antonio
    33207 Gijón (ES)
  • Girela Baena, Lucas
    33207 Gijón (ES)

(74)Representative: Tribalyte Ideas 
Gta. Quevedo 8, 6°2
28015 Madrid
28015 Madrid (ES)

  


(54)DEVICE AND METHOD FOR CONTAINERS CLASSIFICATION


(57) The invention provides a device (1) for containers classification, comprising a first structure (3) and data analysis means (6). The first structure (3) comprises first radiation means (41) suitable for emitting light in the infrared spectrum and first image acquisition means (51) suitable for obtaining images and/or data in the infrared spectrum. The data analysis means (6) is adapted to analyze the images and/or data obtained by the first image acquisition means (51) and providing a classification of the containers (11) piled on the pallet (10). The first image acquisition means (51) have motion means configured to move at least part of the first image acquisition means (51) from a first position where the first image acquisition (51) means allow the platform or pallet (10) enter the first structure and a second position where the first image acquisition means (51) surround the platform or pallet (10). The invention also provides a method for containers classification.




Description

FIELD OF THE INVENTION



[0001] The present invention relates to processes and systems for remote analysis of a group of containers.

BACKGROUND



[0002] The invention is related to the field of analyzing supermarket boxes, of those which are used to contain supplies. These boxes are usually made of plastic and foldable, and are returned to the box supplier in large amount of boxes piled on a support or pallet and wrapped in a transparent film.

[0003] A box provider sends boxes to a supermarket supplier. The supermarket supplier, after using them, returns them to the box provider in order to check them, clean them and repair them, so that they are available for being delivered again.

[0004] Containers are usually manually counted and classified while they are demounted from the pile, and this is usually made one by one. This is not an easy task, mostly when the volume is high, it is costly and subject to human errors. Automation of this process would involve a way of unwrapping the pallet, separating the boxes and counting and classifying them one by one. Further, this should be made when the pallet is provided by the supermarket supplier, in order to check if the pallet contains all the boxes that are supposed to be given back before the supermarket supplier gets away from the delivery place. If a problem with the number and classification of boxes is immediately detected, the supermarket supplier may be warned in the same moment.

[0005] A way of improving this process is therefore sought.

SUMMARY OF THE INVENTION



[0006] The aforementioned problem is solved by a device for containers classification according to claim 1 and a method for containers classification according to claim 12. Particular embodiments of the invention are defined by dependent claims.

[0007] Unless otherwise defined, all terms (including technical and scientific terms) used herein are to be interpreted as is customary in the art. It will be further understood that terms in common usage should also be interpreted as is customary in the relevant art and not in an idealised or overly formal sense unless expressly so defined herein.

[0008] In this text, the term "comprises" and its derivations (such as "comprising", etc.) should not be understood in an excluding sense, that is, these terms should not be interpreted as excluding the possibility that what is described and defined may include further elements, steps, etc.

[0009] In a first inventive aspect, the invention provides a device for containers classification, suitable for classifying containers which are piled on a platform or pallet, the device comprising

conveying means for transporting the pallet

a first structure for pallet analysis, the first structure comprising first radiation means suitable for emitting light in the infrared spectrum and first image acquisition means suitable for obtaining images and/or data in the infrared spectrum; and

data analysis means, such as an image processing software, adapted to analyze the images and/or data obtained by the first image acquisition means and providing a classification of the containers piled on the pallet;

wherein the first image acquisition means have motion means configured to move at least part of the first image acquisition means from a first position where the first image acquisition means allow the platform or pallet enter the first structure and a second position where the first image acquisition means surround the containers.

[0010] When talking about classification, it should be understood in a generic way, as any kind of obtaining features of the boxes to be classified, such as the number of boxes, the type of boxes, the particular model of each box, etc.

[0011] Acquired imagery and/or data may be treated as image or data or a combination of both. There are two different ways of representing and displaying acquired data, either pictorially or digitally, and both are interchangeable as they convey the same information. The term "acquired" does not necessarily mean that the information is stored or retained in any way, it just refers to the fact that a predetermined infrared signal is received or perceived by the first image acquisition means.

[0012] This device solves the problem of automating the process of classifying the containers, since the first acquisition means is configured to read through the plastic film and access to the containers without unwrapping them and separating them.

[0013] In some particular embodiments, the second position is such that the positions of the first image acquisition means define a polyhedron, in such a way that a parallelepiped of 1200 mm high, 1200mm long and 800mm wide is inscriptible inside said polyhedron.

[0014] These dimensions are suitable for the vast majority of pallets which support commercial containers. The positions of the first image acquisition means are suitable for these means to surround the pallet and the containers.

[0015] In some particular embodiments, the conveying means are configured to have a loading height lower than 10 cm. This makes the introduction of the pallet or container much easier, since this container is usually very heavy and raising it to 80 cm, which is the typical height of the loading edge, is very dangerous.

[0016] In some particular embodiments, the device for containers classification further comprises second radiation means suitable for emitting light in the visible spectrum and second image acquisition means suitable for obtaining images in the visible spectrum.

[0017] These second image acquisition means contribute to the analysis with colour data of the containers, since the colour of some containers contributes to their easier identification.

[0018] In some particular embodiments, the first image acquisition means are arranged in four faces which are orthogonal two by two, in such a way that each of the faces comprises a group of first image acquisition means, each group being suitable for covering 5000 px in 1450 mm in width.

[0019] This way of acquiring the image and/or data is particularly advantageous for commercial containers.

[0020] In some particular embodiments, the second image acquisition means are arranged in four faces which are orthogonal two by two, in such a way that each of the faces comprises at least a second image acquisition means being suitable for covering 5000 px in 1450 mm in width.

[0021] This arrangement of the second image acquisition means is consistent with the arrangement of the first image acquisition means, since these second image acquisition means are intended to acquire the same image range of the first image acquisition means.

[0022] In some particular embodiments, the first radiation means and the second radiation means are arranged in four faces which are orthogonal two by two and which form 45° with the faces that comprise the image acquisition means, forming an octagonal structure where faces with radiation means and faces with image acquisition means alternate.

[0023] In this embodiment, the first and second radiation means are arranged to emit radiation in a substantially tangentially direction to the face of the boxes stack, so that glare is minimized.

[0024] In some particular embodiments, each face which comprises radiation means comprises at least two first radiation means and two second radiation means.

[0025] A first radiation means and a second radiation means is intended to emit radiation over a face of the containers stack and the other pair of radiation means is intended to emit radiation on a perpendicular face of the containers stack.

[0026] In some particular embodiments, each of the first and second radiation means comprises lugs to avoid glaring.

[0027] In some particular embodiments, the device further comprises a secondary path intended to be used by packages which have not been correctly classified due to an opaque object which may cover part of the boxes stack.

[0028] This secondary path allows those containers that have been erroneously classified exit by a different place than those that have been correctly classified.

[0029] In some particular embodiments, some image acquisition means are arranged fixed to the first structure and are not affected by the motion means.

[0030] In some particular embodiments, the motion means comprise a rail to allow the sliding of at least some of the first image acquisition means during the data acquisition operation so that they are able to acquire images and/or data in different positions.

[0031] These rails allow that a reduced number of image acquisition means is used, and the way of covering the whole surface of the pallet is that a reduced number of cameras make several shots from different locations. This is a way of reducing the cost of this device.

[0032] In a further inventive aspect, the invention provides a method for containers classification which are piled on a platform or pallet, the method comprising the steps of

acquiring images and/or data of the containers by means of a device according to the first inventive aspect, the images and/or data being acquired in at least four different planes; and

analyzing the images and/or data to provide a classification of containers.



[0033] This method is particularly advantageous to achieve an automation in the process of classifying the containers.

[0034] In some particular embodiments, the data analysis means are configured to analyse at least one of the following criteria:

alphanumeric signal found in each container by OCR;

identification of BIDI codes in each container;

identification of external contours of each container;

identification of internal contours of each container; and/or

identification of the dominant colour of each container.



[0035] These aspects are surprisingly effective in the identification and classification of commercial containers.

[0036] In some particular embodiments, the data analysis means are configured to run algorithms based on neuronal networks, which are intended for machine learning, by using data of previous analysis.

[0037] Machine learning improves the result of this method.

BRIEF DESCRIPTION OF THE DRAWINGS



[0038] To complete the description and in order to provide for a better understanding of the invention, a set of drawings is provided. Said drawings form an integral part of the description and illustrate an embodiment of the invention, which should not be interpreted as restricting the scope of the invention, but just as an example of how the invention can be carried out. The drawings comprise the following figures:

Figure 1 shows a perspective view of a device for containers classification according to the invention.

Figures 2a and 2b shows a detail of two different positions of the device of Figure 1.

Figures 3a and 3b show a front view and a top view of a schematic arrangement of the first image acquisition means in the first structure of a device for containers classification according to the invention.


DETAILED DESCRIPTION OF THE INVENTION



[0039] The example embodiments are described in sufficient detail to enable those of ordinary skill in the art to embody and implement the systems and processes herein described. It is important to understand that embodiments can be provided in many alternate forms and should not be construed as limited to the examples set forth herein.

[0040] Accordingly, while embodiment can be modified in various ways and take on various alternative forms, specific embodiments thereof are shown in the drawings and described in detail below as examples. There is no intent to limit to the particular forms disclosed. On the contrary, all modifications, equivalents, and alternatives falling within the scope of the appended claims should be included. Elements of the example embodiments are consistently denoted by the same reference numerals throughout the drawings and detailed description where appropriate.

[0041] Figure 1 shows a perspective view of a device 1 for containers classification according to the invention. This device 1 is suitable for classifying containers which are piled on a pallet 10. These pallets are usually provided with several containers 11 folded and piled, and the assembly is wrapped in a transparent film.

[0042] This device 1 comprises

conveying means 2 for transporting the pallet 10;

a first structure 3 for pallet analysis, the first structure 3 comprising first radiation means 41, second radiation means 42, first image acquisition means 51 and second image acquisition means 52;

data analysis means 6, such as an image processing software, adapted to analyze the images and/or data obtained by the first image acquisition means and providing a classification of the containers 11 piled on the pallet 10;



[0043] The first radiation means 41 are configured to emit light in the infrared spectrum and the second radiation means 42 are configured to emit light in the visible spectrum. On the other hand, the first image acquisition means 51 are suitable for obtaining images and/or data in the infrared spectrum and the second image acquisition means 52 are suitable for obtaining images in the visible spectrum. The combination of first and second image acquisition means provide a complete picture of the arrangement of the containers 11: the infrared images are not influenced by the transparent film and the colour images provide colour information to the infrared images.

[0044] The conveying means 2 have a loading height lower than 10 cm. This is particularly advantageous when entering the pallet 10 into the device 1.

[0045] As may be seen in this figure, the device 1 also comprises a secondary path 7 intended to be used by packages which have not been correctly classified.

[0046] Figures 2a and 2b shows a detail of two different positions of the device of Figure 1. Figure 2a shows the device of Figure 1 in a first position, where a motor moves some of the first image acquisition means 51 and second image acquisition means 52 to a first position, so that the pallet 10 is allowed to enter the first structure 3. On the other hand, figure 2b shows the device 1 where the motor moves back the first image acquisition means 51 and second image acquisition means 52 to a second position, where the first image acquisition means 51 and second image acquisition means 52 surround the containers 11 and are ready to take the images.

[0047] When the pallet 10 is inside the first structure 3, the following steps are followed:

acquiring images and/or data of the containers by means of the first and second image acquisition means, with the aid of the radiation means, which provides different types of light to enable image acquisition; and

analyzing the images and/or data to provide a classification of containers.



[0048] The radiation is important, since the presence of a protective film around the boxes stack makes it difficult the analysis of raw images. Infrared radiation means provide an important source of radiation which makes this detection and identification easier.

[0049] The data analysis means are configured to analyse at least one of the following criteria:

alphanumeric signal found in each container by OCR;

identification of BIDI codes in each container;

identification of external contours of each container;

identification of internal contours of each container; and/or

identification of the dominant colour of each container.



[0050] The data analysis means are configured to run algorithms based on neuronal networks, which are intended for machine learning, using data of previous analysis as feedback for each analysis step.

[0051] Figures 3a and 3b show a front view and a top view of a schematic arrangement of the first image acquisition means in the first structure.

[0052] Figure 3a shows a front view of one of the four faces where the image acquisition means are arranged. As shown in Figure 2b, the first image acquisition means 51 and second image acquisition means 52 are arranged in four orthogonal faces, while the first radiation means 41 and the second radiation means 42 are also arranged in four orthogonal faces, in such a way that they form an octagonal structure, wherein a side with acquisition means is adjacent to a side with radiation means and vice versa. This octagonal structure is configured to be slid up and down to acquire the images of the pallet and the boxes stacked on it. The positions of the first image acquisition means define a polyhedron, in such a way that a parallelepiped of 1200 mm high, 1200mm long and 800mm wide is inscriptible inside said polyhedron. These are the approximate dimensions of the pallet, so the first and second image acquisition means surround the pallet when images are acquired.

[0053] Each of the four sides which support radiation means comprise four radiation means, two of them are first radiation means 41 and the other two are second radiation means 42. Every radiation means also comprises a lug 8 intended to avoid glaring.

[0054] The radiation means are located in these sides because they provide the radiation tangentially to the boxes stack. Further, these radiation means comprise lugs to avoid the reflection of the radiation emitted in the face of the stack.

[0055] In Figure 3b, different positions of the structure comprising the radiation means 41, 42 and the image acquisition means 51, 52 are shown at the same time. In some embodiments, the device comprises three independent structures, so that all the images are acquired at the same time. In other embodiments, there is only one structure with the radiation means and the image acquisition means, and this structure is moved to acquire the images consecutively.


Claims

1. Device (1) for containers classification, suitable for classifying containers (11) which are piled on a platform or pallet (10), the device (1) comprising

conveying means (2) for transporting the pallet (10);

a first structure (3) for pallet analysis, the first structure (3) comprising first radiation means (41) suitable for emitting light in the infrared spectrum and first image acquisition means (51) suitable for obtaining images and/or data in the infrared spectrum; and

data analysis means (6), such as an image processing software, adapted to analyze the images and/or data obtained by the first image acquisition means (51) and providing a classification of the containers piled on the pallet;

wherein the first image acquisition means (51) have motion means configured to move at least part of the first image acquisition means (51) from a first position where the first image acquisition (51) means allow the platform or pallet (10) enter the first structure and a second position where the first image acquisition means (51) surround the containers (11).
 
2. Device (1) for containers classification according to any of the preceding claims, wherein the conveying means (2) are configured to have a loading height lower than 10 cm.
 
3. Device (1) for containers classification according to any of the preceding claims, further comprising second radiation means (42) suitable for emitting light in the visible spectrum and second image acquisition means (52) suitable for obtaining images in the visible spectrum.
 
4. Device (1) for containers classification according to any of the preceding claims, wherein the first image acquisition means (51) are arranged in four faces which are orthogonal two by two, in such a way that each of the faces comprises a group of first image acquisition means (51), each group being suitable for covering 5000 px in 1450 mm in width.
 
5. Device (1) for containers classification according to claim 4 when depends on claim 3, wherein the second image acquisition means (52) are arranged in four faces which are orthogonal two by two, in such a way that each of the faces comprises at least a second image acquisition means (52) being suitable for covering 5000 px in 1450 mm in width.
 
6. Device (1) for containers classification according to claim 5, wherein the first radiation means (41) and the second radiation means (42) are arranged in four faces which are orthogonal two by two and which form 45° with the faces that comprise the image acquisition means, forming an octagonal structure where faces with radiation means and faces with image acquisition means alternate.
 
7. Device (1) for containers classification according to claim 6, wherein each face which comprises radiation means comprises at least two first radiation means and two second radiation means.
 
8. Device (1) for containers classification according to claim 7, wherein each of the first and second radiation means comprises lugs (8) to avoid glaring.
 
9. Device (1) for containers classification according to any of the preceding claims, further comprising a secondary path (7) intended to be used by packages which have not been correctly classified.
 
10. Device (1) for containers classification according to any of the preceding claims, wherein some first and/or second image acquisition means (51, 52) are arranged fixed to the first structure and are not affected by the motion means (8).
 
11. Device (1) for containers classification according to any of the preceding claims, wherein the motion means (8) comprise a rail to allow the sliding of at least some of the first image acquisition means (51) during the data acquisition operation so that they are able to acquire images and/or data in different positions.
 
12. Method for containers classification which are piled on a platform or pallet, the method comprising the steps of

acquiring images and/or data of the containers by means of a device according to any of the preceding claims, the images and/or data being acquired in at least four different planes; and

analyzing the images and/or data to provide a classification of containers.


 
13. Method for containers classification according to claim 12, wherein the data analysis means are configured to analyse at least one of the following criteria:

alphanumeric signal found in each container by OCR;

identification of BIDI codes in each container;

identification of external contours of each container;

identification of internal contours of each container; and/or

identification of the dominant colour of each container.


 
14. Method for containers classification according to any of claims 12 or 13, wherein the data analysis means are configured to run algorithms based on neuronal networks, which are intended for machine learning, using data of previous analysis.
 




Drawing