Method for detecting the position of containers
The method uses image processing to detect container positions based on predefined size and color, addressing manual annotation complexities and improving packaging line efficiency by reducing blockages.
Patent Information
- Application Number
- FR2022014149
- Authority / Receiving Office
- FR · FR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-12-21
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2042-12-21
AI Technical Summary
Existing container detection systems in packaging lines require complex manual annotation and hardware changes for format or nature changes, leading to frequent blockages and inefficiencies.
A method using image processing to detect container positions based on predefined upper surface size and color, employing local binary patterns and color masking, without the need for manual annotation, to calculate centers of gravity for precise positioning.
Enables reliable container positioning without manual annotation, reducing blockages and enhancing line efficiency by simplifying hardware adjustments.
Smart Images

Figure 00000016_0000 
Figure 00000016_0001 
Figure 00000017_0000
Abstract
Description
Title of the invention: Method for detecting the position of containers TECHNICAL FIELD OF THE INVENTION
[0001] The technical field of the invention is that of container detection, preferably in container packaging lines.
[0002] The present invention relates to a method for detecting the position of containers and in particular for detecting their position via the detection of an upper surface of each container, knowing its size and color. TECHNOLOGICAL BACKGROUND OF THE INVENTION
[0003] In container packaging lines, containers are transported between different processing stations via conveyor belts.
[0004] For the purposes of the present invention, the term "container" encompasses an individual object intended to contain, but not limited to, a fluid, a liquid, powders or granules, in particular of the food or cosmetic type. Such a container may be a bottle or a flask, or a cardboard carton, or even a can. A container may be made of any type of material, in particular plastic, metal or glass. Depending on its shape and material, a container may be rigid or semi-rigid.
[0005] Furthermore, a container may have any type of shape, symmetrical or not, regular or irregular. In addition, a container may have a rounded section, generally circular or ovoid in shape, or a polygonal section, in particular rectangular or square.
[0006] In particular, such a container comprises a bottom which may be flat or substantially flat, or, conversely, having one or more cavities, as is the case, for example, of a can with a concave bottom or a bottle with a so-called "petaloid" bottom.
[0007] In a known manner, within an industrial line, the containers can receive several different successive treatments, such as the manufacture of the container, for example during a plastic injection operation or stretch-blow molding in the case of a plastic bottle, followed by filling and then closing with a cap and labeling. At the end of these treatments, the containers are said to be "finished".
[0008] For the purpose of handling, such containers are packaged in batches.
[0009] Furthermore, each batch comprises a group of several containers, assembled according to a matrix arrangement, usually roughly parallelepiped in shape, often square or rectangular, according to columns and rows. For example, a The usual batch includes six containers in two rows and three columns.
[0010] During these different stages, the containers are transported along the production line, in a direction of movement extending longitudinally, from upstream to downstream, between and within the different stations dedicated to each treatment that the containers must undergo. Such transport can be carried out through at least one conveyor installation.
[0011] Further, the conveying is carried out by moving the containers resting directly at their bottom on the upper face of one or more conveyors. This upper surface is mobile, forming a belt, generally in the form of one or more endless bands wound around at least one motorized winding.
[0012] Along a production line, the containers can be transported in bulk, namely the containers are positioned in a staggered and disorderly manner, in particular against each other, over all or part of the width of a conveyor. For specific processing of the containers, such as packaging, it is necessary to order the bulk flow of containers in several lines, in particular with a view to ensuring their grouping downstream and then batch coating from such a multi-line flow of containers.
[0013] Such an operation is carried out by means of container alignment means, generally taking the form of a funnel positioned above the conveyor, making it possible to receive the bulk flow of said containers at the inlet and to align them in several parallel lines delivered at the outlet. Further, such an alignment means comprises vertical sheets mounted on a frame and spaced so as to form corridors. In addition, the corridors are provided to converge from upstream to downstream, narrowing according to the direction of movement of the containers.
[0014] Thus, at the inlet, the flow of bulk containers separates upon contact with the upstream end of each of the sheets, the containers moving laterally on either side to enter each lane. As they advance, the convergence in each lane forms a bottleneck, requiring the bottles to position themselves one behind the other, in particular by friction upon contact with the walls of the sheets and by rolling and friction between the containers as they advance along each lane.
[0015] In particular, the decreasing width of the corridors is configured as a function of the number of containers to be received at the input, to be channeled to obtain an output flow with lines composed of one or more containers wide, as well as the format of said containers, in particular their diameter.
[0016] A disadvantage of such a funnel aligner lies in the recurring blocking of containers at the inlet and inside the lanes. Indeed, the conveying speed, combined with the narrowing of the lane width, often leads to the posi The frontal operation of two containers which causes them to jam, in particular due to the friction of the products between them and against the walls of the sheets. It is then necessary to stop the conveyor, in order to unblock the products, in particular by manual intervention by an operator: a tedious and precarious operation to reach the containers to be unblocked. This blockage also leads to a loss of production and therefore a loss of efficiency of the line.
[0017] To avoid these container blockages, some solutions propose the detection of the position of the containers by supervised learning. For this, a camera is used and the position of the containers is obtained as output from a machine learning algorithm using a model trained from annotated data. These solutions make it possible, by obtaining the position of the container, to deduce its speed and therefore to predict future blockages. A disadvantage of such solutions is that it requires manual annotation of the position of each container on a plurality of images. When changing containers, packaging lines and / or cameras, new annotated data must be provided to the model, and therefore new annotations must be made by an operator.
[0018] There is therefore a need for a simple solution for obtaining the position of containers in packaging lines that does not require a complex annotation operation when changing the format or nature of the container. Summary of the invention
[0019] The invention offers a solution to the problems mentioned above, by proposing detection of the position of containers by image processing, requiring no annotation and few hardware resources.
[0020] One aspect of the invention relates to a computer-implemented method of detecting the position of a plurality of containers, each container comprising an upper face, the method comprising: • Acquisition of at least one image of the plurality of containers, • Obtaining local binary patterns of the acquired image from a dimension of the upper face of each container of the plurality of containers, the dimension of the upper face of each container being calculated from a predefined upper face size of the containers, • Applying a color mask to the acquired image from a predefined top face color of the containers to obtain a partially masked image, • Combination of the partially masked image and local binary patterns to obtain sets of pixels representing the upper faces of the containers, • Calculation of centers of gravity of the sets of pixels representing the upper faces of the containers of the plurality of containers to obtain the position of each upper face of the containers.
[0021] By means of the invention, a position of the containers of each processed image is obtained from simple data such as the color of the upper surfaces of the containers and the dimension of the upper surfaces of the containers. The method makes it possible to obtain container positions reliably, not requiring annotation of the position of similar containers on similar images unlike the prior art.
[0022] According to additional, non-limiting characteristics, the image is obtained by at least one camera whose field of vision is directed towards the plurality of containers,
[0023] According to additional, non-limiting characteristics, the method further comprises, after the combination of the masked image and the local binary patterns: • At least one expansion of the sets of pixels representing the upper faces of the containers, • At least some erosion of the sets of pixels representing the upper faces of the containers.
[0024] According to one embodiment, the method comprises: • a first expansion of the sets of pixels representing the upper faces of the containers then • a first erosion of the sets of pixels representing the upper faces of the containers then • a second dilation of the sets of pixels representing the upper faces of the containers.
[0025] According to one embodiment, the position of each upper face obtained comprises coordinates of the center of gravity of the upper face in the image.
[0026] According to one embodiment, the combination of the masked image and the local binary patterns is carried out by multiplying a first matrix with a second matrix, the first matrix representing the masked image, the second matrix representing the local binary patterns.
[0027] According to one embodiment, the predefined top surface size and the predefined top surface color are obtained from a human-machine interface, before obtaining local binary patterns and before applying a color mask to the acquired image.
[0028] According to one embodiment, the predefined top surface size and the predefined top surface color are stored in a database.
[0029] According to one embodiment, before calculating the upper surface dimension, the predefined container upper surface size is transformed into a size of container top surface in pixels of the image acquired from a predefined distance between the camera and the top surface of the containers and a predefined camera resolution in pixels.
[0030] Another aspect of the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, cause the latter to implement the method according to the invention.
[0031] Another aspect of the invention relates to a computer-readable recording medium comprising instructions which, when executed by a computer, cause the latter to implement the method according to the invention.
[0032] The invention and its various applications will be better understood by reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES
[0033] The figures are presented for information purposes only and in no way limit the invention. • [Fig.1] shows a schematic representation of a packaging line in which the method according to the invention is implemented, • [Fig.2] shows a schematic representation of the method according to the invention, • [Fig.3] shows a schematic representation of results of the steps of the method according to the invention, • Figures 4a and 4b show schematic representations of containers whose position can be obtained by the method according to the invention, • [Fig.5] shows a schematic representation of a step of the method according to the invention. DETAILED DESCRIPTION
[0034] Unless otherwise specified, the same element appearing in different figures has a single reference.
[0035] [Fig.l] shows a schematic representation of a packaging line in which the method according to the invention can be implemented.
[0036] The method according to the invention, shown schematically in [Fig.2], makes it possible to detect the position of containers, for example in a packaging line.
[0037] Such a packaging line is for example the packaging line 10 of [Fig.l]. The packaging line 10 comprises a belt 12, a set of lanes 13 and a plurality of containers 14. The plurality of containers 14 circulates upright, that is to say resting on their bottom and with a free upper face. The belt 11 makes it possible to move the containers 14 out of and into the lanes 13. The invention applies to any packaging line and is not limited to packaging lines of the type of the packaging line shown in [Fig.l].
[0038] When the invention is applied to the packaging line 10, the packaging line 10 is equipped with a camera 11. The camera 11 is an image capture device, which can be of any type. Preferably, the camera 11 comprises an image sensor, for example of the CCD or CMOS type, and an optical means capable of redirecting the incident light rays towards the image sensor. The camera thus has a field of vision 111 defined in particular by the dimensions of its sensor and by the dimensions and type of its optical means(s). This field of vision 111 is directed towards the containers 14 circulating in or on the packaging line. By "directed towards the containers" is meant that at least a portion of the containers 14, preferably a majority of the containers 14 circulating at a given position are imaged on the sensor of the camera 11.In other words, at each instant, at least a portion of the containers 14, preferably a majority of the containers 14 circulating at a given position, are present on an image obtained by the camera 11 for the given instant.
[0039] The method 20 according to the invention is shown schematically in [Fig. 2]. The method 20 is a computer-implemented method for detecting the position of a plurality of containers.
[0040] The method 20 may be implemented by a computer or by a processor. By "computer-implemented" is meant that the steps, or at least one step, are executed by at least one computer or processor or any other similar system. Thus, steps are performed by the computer, possibly fully automatically or semi-automatically. In examples, the triggering of at least some of the steps of the method may be performed by user-computer interaction. The level of user-computer interaction required may depend on the level of automation intended and balanced against the need to implement the wishes of the user. In examples, this level may be user-defined and / or predefined.
[0041] A typical example of a computer implementation of a method is to execute the method with a system adapted for this purpose. The system may comprise a processor coupled to a memory and a graphical user interface ("GUI"), the memory having recorded thereon a computer program comprising instructions for implementing the method. The memory may also store a database. The memory is any hardware adapted for such storage, possibly comprising several distinct physical parts.
[0042] The method 20 comprises a first step 21 of acquiring at least one image of a plurality of containers 14. The position of the plurality of containers 14 obtained at the end of the method is the position of the containers 14 included in the image, i.e. the containers 14 represented in the form of pixels in the image. The image comprises a plurality of pixels. The image can be obtained, by the processor or the computer implementing the method 20, directly from a camera 11, or from a local or remote database not shown. The image can be acquired by the processor or the computer via a wired or wireless connection. The image is preferably acquired in digital form. The image was previously obtained by the camera 11. The containers 14 present in the image must be seen from a low angle, that is to say that at least one predefined upper surface of the containers 14 must be visible. For this, the camera must be placed at a height relative to the surface on which the containers 14 rest that is greater than the height of the containers 14.
[0043] The acquired image is for example image II represented in [Fig.3], comprising a plurality of containers 14 which are bottles, each bottle comprising a substantially circular white-colored stopper.
[0044] The term "upper surface of the containers" means at least one end of the containers 14. For example, as shown in [Fig. 4a], when the container 14 is a bottle, the upper surface of the bottle is the upper surface 141, i.e., the surface of the bottle cap when the bottle is viewed from above. For example, as shown in [Fig. 4b], when the container 14 is a can, the upper surface of the can is the upper surface 141, i.e., the surface visible when the can is viewed from above. For example, when the container 14 is a brick or a box, the upper surface of the brick or box is the upper surface 141, i.e., the surface visible when the brick or can is viewed from above.
[0045] Similarly, the plurality of containers 14 may be in the form of batches. In such a case, when the containers 14 are in the form of batches 41, the upper surface of the batch 41 is the visible surface when the batch 41 is viewed from above. Alternatively, when the containers 14 are in the form of batches, the upper surface may be the upper surface of each container 14 of the batch 4L
[0046] Generally, the upper surface of a container 14 is the highest surface of the container 14 that is substantially planar or that can be approximated by a substantially planar shape, for example by a solid disc or rectangle. An upper surface is preferably a surface in a horizontal plane, that is to say in a plane that is substantially parallel to the bottom of the container 14 on which the container 14 rests, and therefore substantially parallel to the surface on which the container 14 rests. The “highest” surface is understood to be the highest surface relative to the bottom of the container 14.
[0047] The method then comprises a step 22 of acquiring a size of the upper surface of the containers 14 and a color of the upper surface of the containers 14. This implies that the position of the containers present on the image acquired in step 21 obtained at the end of the method 20 is the position of the containers present on the image and having a surface greater than the acquired size and the acquired color.
[0048] The upper surface size and the upper surface color are acquired, by the processor or the computer implementing the method 20, from a remote and / or local database or directly from a human-machine interface.
[0049] The size of the upper surface of the containers 14 is for example defined in centimeters or millimeters. The size of the upper surface is for example a size of a dimension of the upper surface of the containers 14. A dimension of the upper surface of the containers 14 is for example a diameter of the upper surface 141, for example when the upper surface is the upper surface of a bottle or any container of circular, oval, substantially circular or substantially oval shape, for example a cap diameter when the upper surface is the upper surface of a cap of a bottle 14. A dimension of the upper surface of the containers 14 is for example a width or a length of the upper surface 141, for example when the upper surface is the upper surface of a brick or any container of rectangular or substantially rectangular shape.
[0050] The color of the upper surface of the containers 14 is for example defined in RGB, YCbCr, or any other color format. The color of the upper surface is a majority color of the upper surface, the majority of color being defined in terms of surface. For example, for containers 14 which are bottles with blue caps, the color of the acquired upper surface is the color blue. The color can be a range of colors.
[0051] The size and color of the upper surface of the containers whose position is desired to be obtained are predefined, that is, they are defined before the start of the method 20 and / or before their use by the method 20. For example, the size and color of the upper surface of the containers may be defined by an operator using a human-machine interface such as a touch screen, or a computer comprising a screen and a means for entering data into the computer, for example a keyboard and / or a mouse. After definition by an operator, the size and color of the upper surface of the containers 14 may be stored in a database, accessible by the processor and / or the computer implementing the method 20.Alternatively, the human-machine interface may be directly accessible by the processor and / or computer implementing the method 20 to obtain the upper surface size and color data.
[0052] Thus, thanks to the invention, the change of containers whose position is detected is easy, simple information on color and cap size then being simply entered by an operator and / or automatically retrieved, without requiring long and complex training or data annotation, unlike the art prior.
[0053] Once the size and color of the upper surfaces of the containers have been obtained in step 22, the size of the upper surfaces in pixels must be obtained. When the size of the upper surfaces is already indicated in pixels, the method is continued in step 23. When, on the contrary, the size of the upper surfaces is indicated in a unit other than pixels, the size must be converted into pixels. For this, the conversion is carried out knowing the definition of the image obtained by the camera 11 in pixels (and therefore the resolution of the sensor of the camera 11) and the distance between the camera 11 and the containers 14.
[0054] The method 20 then comprises two steps 23 and 24, which can be performed simultaneously or sequentially, in any order. Steps 23 and 24 make it possible to identify and extract features from the image. Step 23 makes it possible to identify and extract shape features in image II, and step 24 makes it possible to identify and extract color features in image II.
[0055] Step 23 is a step of calculating local binary patterns of the acquired image from a dimension of the upper surface of each container of the plurality of containers, the dimension of the upper surface of each container being calculated from a predefined upper surface size of the containers. When the predefined upper surface size of the containers 14 is equal to the predefined upper surface dimension, the latter is directly used in the calculation of local binary patterns. The predefined upper surface dimension is for example the radius, the diameter, a width or a length of the upper surface.
[0056] The calculation of local binary patterns comprises the use of an algorithm called "Local Binary Pattern" or "LBP". The dimension of the upper surface obtained from the size acquired in step 22 is used to define the neighborhood of each pixel, in number of pixels. The neighborhood is used to compare the luminance level of a central pixel with respect to the luminance level of its neighboring pixels in the neighborhood. For each pixel, called "central", a value is assigned which corresponds to a sequence of numbers (preferably '0' or '1'), each number in the sequence of numbers corresponding to a luminance level of a pixel in the neighborhood with respect to the central pixel. A histogram is then calculated on the neighborhood, of the frequency of appearance of each number or digit (preferably '0' or '1'), that is to say each combination of pixels having a luminance level smaller and larger than the central pixel.This histogram can be considered as a feature vector. The concatenation of the histograms of all the neighborhoods of the image makes it possible to obtain local binary patterns. The invention can be implemented with any algorithm allowing to obtain local binary patterns. A . An example of an image with local binary patterns is shown in [Fig.3], image 12.
[0057] Step 24 is a step of masking the image obtained in step 21 with a color mask. Such a color mask makes it possible to modify the value of all the pixels having a color equal to or close to the predefined color or included in a color range around the predefined color, or included in the predefined color range, and / or to modify the value of all the pixels having a color different from the predefined color not included in a color range around the predefined color or not included in the predefined color range. For example, if the chosen color is white, all the white pixels can remain white and all the non-white pixels can take a similar value different from the value corresponding to the color white. The invention can use any known color mask. The image obtained is said to be masked or “partially” masked, in that only one color is masked.By "masking" we mean the application of a mask, whether it is a mask that reduces the presence of the predefined color or one that attenuates it. A color mask makes it easy to identify all the pixels having the predefined color. An example of a partially masked image obtained is shown in [Fig.3], image 13. .
[0058] The images 12 and 13 obtained are then combined at a step 25. For this, at each step 23 and 24, the image obtained is a digital image in the form of a matrix. The combination of the two images is preferably a multiplication of matrices. Thus, the image 14 resulting from the combination of the images 12 and 13 includes the extracted color and shape information from the image II. This consideration of two types of characteristics makes it possible to better detect the upper surfaces of the containers, and to eliminate at least part of the noise. By "noise" is meant the parts of the image extracted as corresponding to an upper surface of the container but not actually corresponding to it. For example, noise can be created by objects of the same color as the predefined color but not being an upper surface of the container.
[0059] To further reduce noise, after combining images 12 and 13 in step 25 to obtain image 14, an optional step 26 may be implemented.
[0060] Step 26, optional, is shown in [Fig.5] and preferably comprises three sub-steps 261 to 263. Step 26 is applied to sets of pixels from image 14, i.e. from the combination of images 12 and 13.
[0061] In step 26, a succession of erosions and dilations of the sets of pixels from the combined image are carried out. Preferably, the erosions and dilations carried out are as follows: • A first erosion 261, • A first dilation 262, • A second erosion 263.
[0062] This succession of erosions and expansions makes it possible to eliminate small detections, and thus reduce noise. The detection of upper surfaces is then improved.
[0063] "Erosion" means the act of retaining, for an image, only the pixels for which a chosen pattern called a "core" is included in the image. Thus, the image is shrunk.
[0064] "Dilation" means replacing each pixel with a chosen pattern called a "kernel". Thus, the image is enlarged according to the kernel.
[0065] By applying erosions and dilation to each set of pixels from the image 14, the sets of pixels that are too small are deleted, i.e. those having a size smaller than the predefined upper surface size. Indeed, during the first erosion 261, all the pixels of a set of pixels may have been deleted, for example if the core cannot be included in its entirety in the set of pixels. Then, the resulting sets of pixels are dilated in a sub-step 262, to take on a larger size but smaller than their original size. Finally, a second erosion deletes the sets of pixels that are too small. Indeed, during the second erosion 261, all the pixels of a set of pixels may have been deleted, for example if the core cannot be included in its entirety in the set of pixels after dilation.
[0066] This results in a plurality of sets of pixels in the image 15, corresponding to upper surfaces of containers 14. For example, the pixels of the image 15 not belonging to the pixels of the plurality of sets of pixels can be assigned a predetermined value different from the pixels belonging to the pixels of the plurality of sets of pixels, for example the value 0.
[0067] To extract the sets of pixels from the image 15, any algorithm for extracting sets of similar pixels can be used, for example a so-called “label features” algorithm (literally translated as “labeling of characteristics” in French).
[0068] Finally, the method 20 comprises a final step 27 of calculating the centers of gravity of the sets of pixels of the image 15 to obtain the position of each container 14 in the image acquired in step 21. This calculation step makes it possible to obtain coordinates of the containers 14 in the image. To calculate the center of gravity of the sets of pixels, for each set of pixels, the average of the coordinates of the pixels of the set of pixels makes it possible to obtain the center of gravity of the set of pixels.
[0069] The coordinates thus obtained can then be transformed into coordinates relative to the packaging line 10 to position each container 14 relative to the packaging line 10 and not relative to the camera 11. The coordinates relative to the packaging line 10 can then be used to detect and / or predict the formation of blockages in the packaging line, for example by calculating the speed of the containers 14 and their position from successive images at successive times. For this, the method 20 can be repeated every predefined time, for example every 0.2 seconds with a conveyor speed of 300 millimeters per second for a 60 millimeter bottle.
Claims
Claims
1. A computer-implemented method (20) for detecting the position of a plurality of containers (14) in a packaging line, each container (14) comprising an upper surface (141), said plurality of containers (14) circulating upright, i.e. resting on their bottom and with said upper surface (141) free, the method comprising: - Acquiring (21) at least one image (II) of the plurality of containers (14), - Obtaining (23) local binary patterns (12) of the image (II) acquired from a dimension of the upper surface (141) of each container (14) of the plurality of containers (14), the dimension of the upper surface (141) of each container (14) being calculated from a predefined upper surface size (141) of the containers (14),- Applying (24) a color mask to the image (II) acquired from a predefined upper surface color (141) of the containers (14) to obtain a masked image (13), - Combining (25) the partially masked image (13) and the local binary patterns (12) to obtain sets of pixels representing the upper surfaces (141) of the containers (14), - Calculating (27) centers of gravity of the sets of pixels representing the upper surfaces (141) of the containers (14) of the plurality of containers (14) to obtain the position of each upper surface (141) of the containers (14).,
2. Method (20) according to claim 1 according to which the image (II) is obtained by at least one camera (11) whose field of vision (111) is directed towards the plurality of containers (14).
3. Method (20) according to one of the preceding claims further comprising, after the combination (25) of the masked image (13) and the local binary patterns (12): - At least one expansion (261, 263) of the sets of pixels representing the upper surfaces (141) of the containers (14), the dilation consisting of replacing each pixel with a chosen pattern called a “core”, - At least one erosion (262) of the sets of pixels representing the upper surfaces (141) of the containers (14), the erosion consisting of retaining, for an image, only the pixels for which said chosen pattern called a “core” is included in the image.
4. Method (20) according to claim 3 comprising: - a first expansion (261) of the sets of pixels representing the upper surfaces (141) of the containers (14) then - a first erosion (262) of the sets of pixels representing the upper surfaces (141) of the containers (14) then - a second expansion (263) of the sets of pixels representing the upper surfaces (141) of the containers (14).
5. Method (20) according to one of the preceding claims according to which the position of each upper surface (141) obtained comprises coordinates of the center of gravity of the upper surface (141) in the image.
6. Method (20) according to one of the preceding claims, according to which the combination (25) of the masked image (13) and the local binary patterns (12) is carried out by multiplication of a first matrix with a second matrix, the first matrix representing the masked image (13), the second matrix representing the local binary patterns (12).
7. Method (20) according to one of the preceding claims according to which the predefined upper surface size (141) and the predefined upper surface color (141) are obtained from a human-machine interface, before obtaining (22) local binary patterns and before applying (23) a color mask to the acquired image (II).
8. Method (20) according to one of the preceding claims, wherein the predefined upper surface size (141) and the predefined upper surface color (14) are stored in a database.
9. Method (20) according to one of the preceding claims according to which, before calculating the dimension of the top surface (141), the predefined top surface size (141) of container (14) is transformed into the top surface size (141) of container (14) in pixels of the image acquired from a predefined distance between the camera (11) and the top surface of the containers (14) and a predefined resolution of the camera (11) in pixels.
10. Computer program product comprising instructions which, when the program is executed by a computer, cause the latter to implement the method (20) according to one of claims 1 to 8.
11. A computer-readable recording medium comprising instructions which, when executed by a computer, cause the computer to carry out the method (20) according to one of claims 1 to 8.
