SYSTEM AND METHOD FOR IDENTIFICATION AND/OR SORTING OF OBJECTS

Abstract

The invention relates to a method for identifying and/or sorting objects, in particular for recycling materials, comprising the steps of: Linking at least one first object type to object identity information via a reference object type property uniquely identifying the first object type; performing at least one learning phase for teaching at least one KI algorithm, the learning phase comprising analyzing at least one object having the reference object type property for an object property; establishing a correlation between the object identity information and the at least one object property, the correlation comprising associating the at least one analyzed object with the first object type; analyzing at least one object for at least one object property; and calculating an object identity of the object to the first object type using the at least one KI algorithm. The invention further relates to a system for identifying and/or sorting objects based on artificial intelligence technologies.

Claims

1-14. (canceled)

15. A method for the identification and/or sorting of objects, in particular for the recycling of materials, comprising: Linking at least one first object type to object identity information via a reference object type property uniquely identifying the first object type; Performing at least one learning phase for teaching at least one AI algorithm, the learning phase comprising analyzing at least one object having the reference object type property for an object property; Establishing a correlation between the object identity information and the at least one object property, the correlation comprising associating the at least one analyzed object with the first object type; Analyzing at least one object for the at least one object property; and Calculating an object identity of the object to the first object type using the at least one AI algorithm.

16. The method of claim 15, wherein the reference object type property is a fluorescent code.

17. The method of claim 15, wherein the reference object type property comprises XRF (X-ray fluorescence analysis) codes and/or magnetic codes and/or particle codes and/or electronic data and/or watermarks and/or bar codes and/or QR codes and/or symbols and/or part numbers and/or design elements and/or native object properties, such as the chemical material composition, and/or the color and/or the shape and/or the size and/or the surface structure of the at least one analyzed object.

18. The method of claim 15, further comprising transmitting single object identities computed by at least two different AI algorithms to a voting algorithm for computing a combined object identity.

19. The method of claim 15, further comprising sorting the objects according to the calculated object identity or combined object identity.

20. The method of claim 15, wherein no further learning phase is performed after completion of the learning phase.

21. The method of claim 15, wherein the calculation of the object identity is performed by analyzing a reference object type property after completion of the learning phase.

22. The method of claim 15, wherein at least one reference object with at least one reference object type property is provided for the learning phase.

23. A method for teaching at least one AI algorithm for recycling materials, comprising the steps of Determining at least one object property of a first object type as a reference object type property for a learning phase; Analyzing object properties of at least one object, wherein the at least one object is checked for the presence of the at least one reference object type property; Calculating an object identity of the at least one object to the first object type based on at least one object property; Teaching at least one AI algorithm using the at least one object property, wherein a correlation is established between the at least one analyzed object property and the object identity.

24. A system for identifying and/or sorting objects for recycling materials, which comprises: a detection system with at least one detection module, wherein the at least one detection module is adapted to analyze object properties of objects; a computer system, the computer system being adapted to process the analyzed object properties to compute object identities of the analyzed objects, the computer system including a system for storing, executing, and training one or more AI algorithms, wherein said training of the AI algorithms is based on one or more object properties; wherein the AI algorithms are adapted to compute an object identity for analyzed objects.

25. The system of claim 24, wherein the computer system includes a voting algorithm configured to combine object identities calculated by all algorithms to calculate a combined object identity.

26. The system of claim 25, wherein the voting algorithm is arranged to perform a weighting of the object identities calculated by all algorithms.

27. The system of claim 25, further comprising a sorting device arranged to sort the objects according to the object identity calculated by the computer system.

28. The system of claim 25, wherein analyzing the object properties comprises checking a reference object type property.

Description

EXAMPLES

Example 1

[0101] Two types of plastic bottles A and B are to be identified by their bottle shape. The bottle shape is detected by a detection module with camera system. The bottles can be present in the detection module with random orientation, which makes automatic image recognition difficult.

[0102] The identification of the bottles is performed by means of AI technology. For autonomous training of the system by the object property bottle shape the bottles are marked with two different fluorescence codes a and b. The fluorescence codes a and b act as a reference object type property. The fluorescence codes a and b are stored in a database together with the associated object identities A and B.

[0103] For the learning phase, a plurality of plastic bottles A and B are analyzed autonomously by the system. The learning phase includes the analysis of the object property bottle shape by capturing images of the bottles, the presence check of the reference object type properties code a and code b, the assignment of the object identity bottle A to bottles with code a, the assignment of the object identity bottle B to bottles with code b and the adaptation of the algorithm which processes the images to detect patterns and regularities in the images and establish a correlation between the captured images and the object identity.

[0104] Once the learning phase is complete, the system can calculate the identity of the plastic bottles based on the captured images. The fluorescent codes used in the learning phase no longer need to be present on the bottles for this purpose.

[0105] When a sorting unit is integrated, the sorting unit can sort bottles A and B into different containers.

Example 2

[0106] Two cosmetic bottles A and B have different label designs. The bottles are to be identified by their optical design. The design is detected by a detection module with a camera system.

[0107] The identification of the bottles is performed by means of AI technology. For autonomous training of the system by the object property optical design, the bottles are marked with two different fluorescence codes a and b. The fluorescence codes a and b act as a reference object type property. The fluorescence codes a and b are stored in a database together with the associated object identities A and B.

[0108] For the learning phase, a plurality of bottles A and B are analyzed autonomously by the system. The learning phase includes the analysis of the object property optical design by capturing images of the bottles, the presence check of the reference object type properties code a and code b, the assignment of the object identity bottle A to bottles with code a, the assignment of the object identity bottle B to bottles with code b and the adaptation of the algorithm which processes the images to detect patterns and regularities in the designs and establish a correlation between the captured designs and the object identity.

[0109] Once the learning phase is complete, the system can calculate the identity of the bottles based on the captured images. The fluorescence codes used in the learning phase no longer need to be present on the bottles for this purpose.

[0110] When a sorting unit is integrated, the sorting unit can sort bottles A and B into different containers.

Example 3

[0111] Two types of packaging A and B, on which labels A and B are located, are to be identified via water marks integrated into the labels. It should also be possible to distinguish packages A and B from package C, where package C does not contain a watermark in the label. The watermarks are detected by a detection module with a camera system. The packages and thus labels may be present with random orientation in the detection module. Furthermore, the labels may be dirty and mechanically deformed. These factors complicate the automatic detection of the watermarks.

[0112] The identification of the labels is performed by means of AI technology. For autonomous training of the system by the object property water mark, the labels A and B are marked with two different fluorescence codes a and b.

[0113] The fluorescence codes a and b act as a reference object type property. The fluorescence codes a and b are stored in a database together with the associated object identities A and B. Label C does not receive a fluorescence code, so it does not contain a reference object type property.

[0114] For the learning phase, a large number of packages with labels A, B and C are analyzed autonomously by the system. The learning phase includes the analysis of the object property watermark by analyzing the labels, the presence check of the reference object type properties code a and code b, the assignment of the object identity label A to labels with code a, the assignment of the object identity label B to labels with code b and the adaptation of the algorithm which processes the watermarks to detect patterns and regularities in the watermarks and establish a correlation between the detected watermarks and the object identity. Also for labels C, the analysis of the object property watermark is performed by the detection module and the presence check of the reference object type property fluorescent code is performed. Since the reference object type property is not found, objects with label C receive the object identity unknown. Therefore, for labels C, the algorithm learns the correlation between the object identity unknown and labels without watermark.

Example 4

[0115] Two types of packaging A and B are to be identified by labels with fluorescent codes applied to them. The fluorescence code is detected by a detection module with spectrometer.

[0116] The identification of the packages is performed by means of AI technology. For autonomous training of the system by the object property fluorescence code, the different geometries of the packages A and B are used as reference object type property. The geometries a and b together with the corresponding object identities A and B are stored in a database.

[0117] For the learning phase, a plurality of packages A and B are analyzed autonomously by the system. The learning phase includes the analysis of the object property fluorescence code by spectrometer analysis, the presence check of the reference object type properties geometry a and geometry b, the assignment of the object identity packaging A to packaging with geometry a, the assignment of the object identity packaging B to packaging with geometry b and the adaptation of the algorithm which processes the fluorescence spectra to detect patterns and regularities in the spectra and establish a correlation between the acquired spectra and the object identity.

[0118] Once the learning phase is complete, the system can calculate the identity of the packages based on the detected fluorescence codes. The packages no longer have to have the packaging geometries used in the learning phase.

[0119] When a sorting unit is integrated, the sorting unit can sort packages A and B into different containers.

Example 5

[0120] Packaging contains a luminescent marker in the base material of the packaging. The packaging shows varying degrees of contamination. The influence of the contamination on the variance of the emission spectrum of the luminescence marker is to be analyzed.

[0121] A variety of packages is analyzed by the system. The analysis of the object property emission spectrum is performed by spectrometer analysis. As results the variances of emission intensity, emission maxima (wavelengths with maximum emission), and half-widths are obtained. The obtained results can now be used to adjust the spectrometer sensor technology.

Example 6

[0122] A wide variety of detection modules are used in a sorting system. For example, electrical conductivity, the IR reflectance spectrum, watermarks and fluorescence codes are detected. Two types of objects A and B are to be identified. The identification of the objects will be performed using AI technology. For autonomous training of the system, object A is marked with fluorescence codes a and object B is marked with fluorescence code b. The fluorescence codes a and b act as a reference property. The fluorescence codes a and b together with the corresponding object identities A and B are stored in a database.

[0123] The learning of the AI technology is to be carried out during operation. For the learning phase, objects A and B are mixed among other objects transported through the sorting system. The objects are analyzed autonomously by the system. The learning phase includes the analysis of the object properties electrical conductivity, IR reflectance spectrum and watermark, the presence check of the reference properties code a and code b, the assignment of the object identities A to objects with fluorescence code a and B to objects with fluorescence code b, the adaptation of the algorithms to detect patterns and regularities in the analyzed object properties and the establishment of a correlation between the detected properties and the object identity.

[0124] Once the learning phase is complete, the system can calculate the identity of the objects based on the detected object properties electrical conductivity, IR reflectance spectrum and watermark, and sort the objects A and B. The fluorescence codes used in the learning phase no longer need to be present on the objects for this purpose.

[0125] To test the object identification and sorting, the matching of the captured reference properties with the reference properties stored in the database and linked to object identities is deactivated. Subsequently, it is checked whether objects A and B are still correctly identified and sorted. Furthermore, objects A and B without fluorescence codes can be processed by the sorting system and their identification and sorting checked.

[0126] The foregoing description of the embodiments has been provided for purposes of description. It is not intended to be exhaustive or to limit the disclosure. Individual elements or features of a particular embodiment are generally not limited to that particular embodiment, but, where applicable, are interchangeable and can be used in a selected embodiment, even if not specifically shown or described. The same may also be varied in many ways. Such variations are not to be regarded as a departure from the disclosure, and all such modifications are intended to be included within the scope of the disclosure.