Method for classifying elements

Abstract

Element classification method The present invention relates to an element classification method comprising: the collection of a set of three-dimensional models of elements, the model of each element being labeled with a classification of the element, for each three-dimensional model collected, the generation of a series of images of the element, each image of the element being labeled with the classification of the element, the images generated for the set of elements forming a training image base, the training of an element classification model according to a learning technique applied to the training image base to obtain a trained classification model, the exploitation of the trained classification model comprising: the reception of an image of an element to be classified, and the determination of a classification of the element by the trained classification model.the implementation of an action relating to the element based on the classification obtained for the element. Figure for the abbreviation: 3.

Description

Title of the invention: Method for classifying elements

[0001] The present invention relates to a method for classifying elements. The present invention also relates to an associated computer program product.

[0002] In the industrial field, many elements, such as industrial parts or office elements, are handled by the different actors.

[0003] However, when one of these elements needs to be renewed, these players do not necessarily turn to a supplier approved by their company due to a lack of awareness of the existence of such suppliers.

[0004] Also, a tool allowing a supplier to be associated with such elements would be of interest.

[0005] For this purpose, in the field of research, classification tools based on neural networks exist. Such neural networks are notably trained from a database of element images, called a training image database.

[0006] However, such classification tools are not suitable for the industrial context. In particular, to obtain a reliable classification tool, current tools have been trained on a multitude of images of elements taken under different conditions (angles, distances, brightness, textures).

[0007] However, in an industrial context, the production of such a training image base with sufficient quality proves to be very complex and time-consuming to implement, and is therefore not relevant in terms of feasibility.

[0008] There is therefore a need for a method for reliably classifying elements in an industrial context.

[0009] To this end, the present description relates to a method for classifying elements comprising:

[0010] a. a phase of collecting a set of three-dimensional models of elements, the model of each element being labeled with a classification of the element, the collection phase being implemented by computer, b. for each three-dimensional model collected, a phase of generating a series of images of the element, each image of the element being labeled with the classification of the element, the images generated for all the elements forming a training image base, the generation phase being implemented by computer, c. a training phase of an element classification model according to a learning technique applied to the training image base to obtain a trained classification model, the training phase being computer implementation, d. an exploitation phase of the trained classification model, the exploitation phase being implemented by computer and comprising: i. a step of receiving an image of an element to be classified, and ii. a step of determining a classification of the element by the trained classification model. e. a phase of implementing an action relating to the element based on the classification obtained for the element.

[0011] According to particular embodiments, the control method comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations:

[0012] - the method comprises images of the element taken from different angles and / or at different distances from the element;

[0013] - the method comprises, upstream of the generation phase, a processing phase of the three-dimensional model of each element, the processing phase advantageously including the filtering of the texture and / or color characteristics present on the model;

[0014] - the implementation phase includes the purchase of an item having the same classi fication that the classified element during the exploitation phase, the implementation phase being carried out by computer, preferably automatically;

[0015] - the implementation phase includes the generation of an alert based on the classification obtained for the element, the implementation phase being carried out by computer, preferably automatically;

[0016] - the implementation phase includes the launch of a control application of the element based on the classification obtained for the element, the implementation phase being carried out by computer, preferably automatically;

[0017] - during the exploitation phase, each classification is associated with a probability representative of a level of confidence in the classification;

[0018] - the image of each element to be classified during the exploitation phase is obtained via a sensor, for example a camera;

[0019] - the elements are chosen from the list consisting of: industrial equipment, industrial equipment parts, personal protective equipment, parts of personal protective equipment and office supplies.

[0020] The present description also relates to a computer program product comprising a readable information medium, on which is stored a computer program comprising program instructions, the computer program being loadable onto a data processing unit and adapted to cause the implementation of a method as previously described when the program computer is implemented on the data processing unit.

[0021] The present description also relates to a readable information medium on which a computer program product as previously described is stored.

[0022] Other characteristics and advantages of the invention will appear on reading the following description of embodiments of the invention, given by way of example only and with reference to the drawings which are:

[0023] [fig. 1], [fig.l], a schematic view of an example of a computer allowing the implementation of a method for classifying elements,

[0024] [fig.2], [fig.2], a flowchart of an example of implementation of a method of classification of elements,

[0025] [fig.3], [fig.3], a schematic representation of an example of implementation of the phase of generating a series of images of an element from its three-dimensional model, and

[0026] [fig.4], [fig.4], a schematic representation of an example of a series of images generated for each element during the generation phase.

[0027] A calculator 10 and a computer program product 12 are illustrated in FIG. 1.

[0028] The calculator 10 is preferably a computer.

[0029] More generally, the computer 10 is an electronic computer capable of manipulating and / or transforming data represented as electronic or physical quantities in computer 10 registers and / or memories into other similar data corresponding to physical data in memories, registers or other types of display, transmission or storage devices.

[0030] The calculator 10 interacts with the computer program product 12.

[0031] As illustrated by [fig.l], the computer 10 comprises a processor 14 comprising a data processing unit 16, memories 18 and an information medium reader 20. In the example illustrated by [fig.l], the computer 10 comprises a keyboard 22 and a display unit 24.

[0032] The computer program product 12 comprises an information medium 26.

[0033] The information medium 26 is a medium readable by the computer 10, usually by the data processing unit 16. The readable information medium 26 is a medium adapted to memorize electronic instructions and capable of being coupled to a bus of a computer system.

[0034] By way of example, the information medium 26 is a floppy disk or flexible disk (from the English name “Floppy disk”), an optical disk, a CD-ROM, a magneto-optical disk, a ROM memory, a RAM memory, an EPROM memory, an EEPROM memory, a magnetic card or an optical card.

[0035] The computer program 12 comprising program instructions is stored on the information medium 26.

[0036] The computer program 12 is loadable onto the data processing unit 16 and is adapted to cause the implementation of a method for classifying elements E1, ..En when the computer program 12 is implemented on the processing unit 16 of the computer 10

[0037] The elements El, ..., En to be classified are, for example, chosen from the list consisting of: industrial equipment, parts of industrial equipment, personal protective elements (helmets, vests, shoes, suits), parts of personal protective elements and office supplies. The term "parts" is understood to mean the different components of the equipment.

[0038] The operation of the calculator 10 will now be described with reference to [fig.2], which schematically illustrates an example of implementation of a method for classifying elements and to figures 3 and 4 which illustrate in more detail certain phases of this method.

[0039] The classification method comprises a phase 100 of collecting a set of three-dimensional models Ml,..., Mn of elements El,..., En. Such models Ml,..., Mn are in particular computer models of the elements El,..., En. Such models Ml,..., Mn have, for example, been obtained by a technique or via computer-aided design software.

[0040] The three-dimensional model Ml,..., Mn of each element El,..., En is labeled with a classification of the element El,..., En. The classification of the element El,..., En is a category to which the element El,..., En is assigned. The category is, for example, the nature of the element El,..., En, an approved supplier for the renewal of the element El,..., En or the serial number of the element El,..., En. The classification of the element El,..., En has, for example, been determined by an operator.

[0041] The collection phase 100 is implemented by the computer 10 in interaction with the computer program product 12, that is to say is implemented by computer.

[0042] Optionally, the classification method comprises a phase 110 of processing the three-dimensional model Ml,..., Mn of each element El,..., En. The processing phase makes it possible to simplify the model Ml,..., Mn by filtering characteristics specific to an element El,..., In particular, but not representative of the class of the element El,..., En.

[0043] For example, the processing phase includes the filtering of the texture and / or color characteristics present on the model Ml,..., Mn of the element El,..., En. Thus, this makes it possible to subsequently generate an independent training database. depending on the textures and / or colors of the elements El, En considered which are characteristics having a high variability (object under corrosion, repainted object).

[0044] The processing phase 110 is implemented by the computer 10 in interaction with the computer program product 12, that is to say is implemented by computer.

[0045] The classification method comprises, for each collected three-dimensional model Ml,..., Mn, a phase 120 of generating a series of images IM1,..., IMn of the element El,..., En. Each image IM1,..., IMn of the element El,..., En is labeled with the classification of the element El,..., En (the same as that of the associated three-dimensional model Ml,..., Mn).

[0046] Advantageously, the series of images IM1, ..., IMn of each element El, ..., En comprises images IM1, ..., IMn of the element El, ..., En taken from different angles and / or at different distances from the element El, ..., En. The successive images IM1, ..., IMn are, for example, acquired with a rotation angle step and / or a distance step from the element El, ..., En.

[0047] The generation phase 120 is similar to a “shooting” of the element El, ..., En. In particular, [fig.3] illustrates a model Ml of an element (clamp) which is rotated to obtain images IM1, IM2, ..., IMn of the element from different angles and at different distances.

[0048] The images IM1, ..., IMn generated for all the elements El, ..., En form a training image base. The example in [fig.4] illustrates in particular the series of images IM1, ..., IMn obtained for four elements El, E2, E3, E4. The number of images IM1, ..., IMn taken for each element depends, for example, on the size, surface area and complexity of the element.

[0049] The training base obtained is, for example, stored in a memory 18 of the computer 10.

[0050] The generation phase 120 is implemented by the computer 10 in interaction with the computer program product 12, that is to say is implemented by computer.

[0051] The classification method comprises a phase 130 of training a classification model of elements El, ..., En according to a learning technique applied to the training image base to obtain a trained classification model.

[0052] The classification model is, for example, a neural network. The learning technique makes it possible to configure the neural network as it is learned based on training images.

[0053] In one example, a portion of the training image database is used to configure the neural network and the other portion to validate the configuration.

[0054] The training phase 130 is implemented by the computer 10 in interaction with the computer program product 12, that is, is implemented by computer.

[0055] The configuration method comprises an exploitation phase 140 of the trained classification model.

[0056] The exploitation phase 140 notably comprises the reception of an image IM1, ..., IMn of an element El, ..., En to be classified, and a step of determining a classification of the element El, ..., En by the trained classification model.

[0057] Advantageously, the image IM1, ..., IMn of the element El, ..., En to be classified is obtained via a measurement by a sensor, such as a camera.

[0058] Advantageously, each classification is associated with a probability representative of a level of confidence in the classification. This allows the operator to verify a classification associated with an unsatisfactory probability, typically lower than a predetermined threshold (for example lower than 85%).

[0059] The operating phase 140 is implemented by the computer 10 in interaction with the computer program product 12, that is to say is implemented by computer.

[0060] The classification method comprises a step 150 of implementing an action relating to the element El, ..., En as a function of the classification obtained for the element El, ..., En. The implementation phase is, for example, carried out by computer, preferably automatically. The action implemented depends in particular on the type of elements El, ..., En and the intended application.

[0061] For example, the implementation phase includes the purchase of an element El, ..., En having the same classification as the element El, ..., En classified during the operation phase. The purchase is, for example, carried out automatically based on the classification obtained. Alternatively, the purchase is validated by an operator before being carried out. Such an action is part, for example, of a purchase recommendation process ("guided buying") within a company, for example to renew parts of equipment or office supplies. In an example scenario, a machine breaks down and must be replaced. An operator in the field is not aware that a contract with local suppliers exists. Therefore, without the classification model, the operator will renew the machine with a non-contractual supplier.On the other hand, with the classification model, the operator provides a photo of the defective machine as input to the model, and the model provides information (number for example) as output allowing the link to be made with the contractual supplier.

[0062] In another example, the implementation phase comprises the generation of an alert based on the classification obtained for the element El, ..., En. Such an action is part, for example, of the monitoring of an industrial site or people. The alert allows, for example, to alert an operator when an unauthorized El, ..., En element is identified on the site or when an individual is not equipped with a mandatory El, ..., En protective element (helmet).

[0063] In yet another example, the implementation phase comprises the launching of an application for controlling the element El, ..., En based on the classification obtained for the element El, ..., En. This thus makes it possible to launch automatic piloting of the element El, ..., En. Such an action is, for example, carried out by an autonomous robot.

[0064] Thus, the present method makes it possible to reliably classify elements in an industrial context. Indeed, the classification model used was trained on a plurality of images of elements obtained from a three-dimensional model of labeled elements. The images obtained are synthetic images obtained during a “shoot” of the three-dimensional model. Obtaining the training image base thus does not require any acquisition of real images.

[0065] The method of obtaining the image base is thus simple to implement and not very time-consuming compared to actual image capture. It is therefore suitable for an industrial context.

[0066] Those skilled in the art will understand that the previously described embodiments and variants may be combined to form new embodiments provided that they are technically compatible.

Claims

Claims

1. Method for classifying elements (El, En) comprising: a. a collection phase of a set of three-dimensional models (Ml,..., Mn) of elements (El,..., En), the model (Ml,..., Mn) of each element (El,..., En) being labeled with a classification of the element (El,..., En), the collection phase being implemented by computer, b. for each three-dimensional model (Ml,..., Mn) collected, a phase of generating a series of images (IM1,..., IMn) of the element (El,..., En), each image (IM1,..., IMn) of the element (El,..., En) being labeled with the classification of the element (El,..., En), the images (IM1,..., IMn) generated for all the elements (El,..., En) forming a training image base, the generation phase being implemented by computer, the method comprising, upstream of the generation phase, a phase of processing the three-dimensional model (Ml,..., Mn) of each element (El,..., En), the processing phase advantageously comprising the filtering of the texture and / or color characteristics present on the model (Ml,..., Mn) of the element (El,..., En), c. a training phase of an element classification model (El, ..., En) according to a learning technique applied to the training image base to obtain a trained classification model, the training phase being implemented by computer, d. an exploitation phase of the trained classification model, the exploitation phase being implemented by computer and comprising: i. a step of receiving an image (IM 1, ..., IMn) of an element (El, ..., En) to be classified, and ii. a step of determining a classification of the element (El, ..., En) by the trained classification model. e. a phase of implementation of an action relating to the element (El, ..., En) based on the classification obtained for the element (El, ..., En).

2. A method according to claim 1, wherein the series of images (IM1, IMn) of each element (El, ..En) comprises images of the element (El, ..., En) taken from different angles and / or at different distances from the element (El, ..., En).

3. Method according to claim 1 or 2, wherein the implementation phase comprises the purchase of an element (El, ..., En) having the same classification as the element (El, ..., En) classified during the exploitation phase, the implementation phase being carried out by computer, preferably automatically.

4. Method according to claim 1 or 2, in which the implementation phase comprises the generation of an alert according to the classification obtained for the element (El, ..., En), the implementation phase being carried out by computer, preferably automatically.

5. Method according to claim 1 or 2, in which the implementation phase comprises launching an application for controlling the element (El, ..., En) according to the classification obtained for the element (El, ..., En), the implementation phase being carried out by computer, preferably automatically.

6. Method according to any one of claims 1 to 5, in which during the exploitation phase, each classification is associated with a probability representative of a level of confidence in the classification.

7. Method according to any one of claims 1 to 6, in which the image (IM1, ..., IMn) of each element (El, ..., En) to be classified during the exploitation phase is obtained via a sensor, for example a camera.

8. A method according to any one of claims 1 to 7, wherein the elements (El, ..., En) are chosen from the list consisting of: industrial equipment, parts of industrial equipment, personal protection elements, parts of personal protection elements and office supplies.

9. A computer program product comprising program instructions recorded on a computer-readable medium, for executing a method according to any one of claims 1 to 8 when the computer program is executed on a computer.

Images