Computer-implemented method for classifying waste

1 . A computer-implemented method for classifying waste into predefined waste categories K 1 , K 2 , . . . , K n using an image recording unit and a data processing unit, wherein the method comprises:

recording, by way of the image recording unit, an image of unseparated waste that is to be separated, the unseparated waste comprising waste objects;

detecting, by way of a neural image recognition network that is connected to the data processing unit or integrated into the data processing unit, the waste objects that are depicted in the image; and

classifying at least one of the waste objects into one of the predefined waste categories K 1 , K 2 , . . . , K n ;

wherein the waste objects are classified into one of the predefined waste categories K 1 , K 2 , . . . , K n on the basis of a geographical location “P” the geographical location of the image recording unit or the unseparated waste, wherein the predefined waste categories comprise at least residual waste, organic waste, paper, plastic, glass, and hazardous waste, wherein the geographical location “P” is provided by a location sensor that is connected to the data processing unit, and wherein a database, which is connected to the data processing unit or integrated into the data processing unit and in which a relationship between the location “P”, predefined waste objects, and the predefined waste categories K 1 , K 2 , . . . , K n is stored, is used to classify the waste objects into one of the predefined waste categories K 1 , K 2 , . . . , K n .

2 . The method according to claim 1 , wherein a neural waste classification network, which is connected to the data processing unit or integrated into the data processing unit and has been trained with training data mapping a relationship between the location “P”, predefined waste objects and the waste categories K 1 , K 2 , . . . , K n , is used to classify the waste objects into one of the predefined waste categories K 1 , K 2 , . . . , K n .

3 . The method according to claim 1 , wherein the image recognition network or the data processing unit creates from the image sections displaying detected waste objects and uses these image sections to classify the detected waste objects.

4 . The method according to claim 1 , wherein the geographical location “P” is provided by a database connected to the data processing unit via a communication unit, wherein the database is located on a server on the Internet, on the basis of user data stored in the data processing unit.

5 . The method according to claim 1 , wherein the data processing unit creates a modified image comprising the image that highlights the waste objects detected in the image and in that the modified image is displayed on a user interface, particularly on a touch screen and the user is prompted to select at least one of the highlighted waste objects in the modified image, whereupon the classified waste category K 1 , K 2 , . . . , K n of the selected waste object is displayed on the user interface.

6 . The method according to claim 5 , wherein, by querying a database connected to the data processing unit via a communication unit, the database being located on a server on the Internet, information I relating to the disposal of the waste object is determined on the basis of the waste categories K 1 , K 2 , . . . , K n determined for the waste object selected by the user and the location “P” and this information I is displayed on the user interface.

7 . The method according to claim 1 , wherein the data processing unit selects a local provider “A” for the disposal of the waste object selected by the user by querying a database connected to the data processing unit via a communication unit, the database being located on a server on the Internet, and displays the determined provider on the user interface.

8 . The method according to claim 7 , wherein the data processing unit ends an electronic order via a communication unit to the provider “A” for the disposal of the classified waste object.

9 . A non-transitory computer-readable storage medium comprising instructions which cause an electronic device comprising an image recording unit and a data processing unit to employ a method comprising

recording, by way of the image recording unit, an image of unseparated waste that is to be separated, the waste comprising waste objects;

detecting, by way of a neural image recognition network that is connected to the data processing unit or integrated into the data processing unit, the waste objects depicted in the image; and

classifying at least one of the waste objects into one of the predefined waste categories K 1 , K 2 , . . . , K n ;

wherein the waste objects are classified into one of the predefined waste categories K 1 , K 2 , . . . , K n on the basis of a geographical location “P” of the image recording unit or the unseparated waste, wherein the predefined waste categories comprise at least residual waste, organic waste, paper, plastic, glass, and hazardous waste,

wherein the geographical location “P” is provided by a location sensor that is connected to the data processing unit,

wherein a database, which is connected to the data processing unit or integrated into the data processing unit and in which a relationship between the location “P”, predefined waste objects and the predefined waste categories K 1 , K 2 , . . . , K n is stored, is used to classify the waste objects into one of the predefined waste categories K 1 , K 2 , . . . , K n .

10 . A device for classifying waste comprising an image recording unit and a data processing unit wherein the device is configured to employ a method comprising

recording, by way of the image recording unit, an image of unseparated waste that is to be separated, the waste comprising waste objects;

detecting, by way of a neural image recognition network that is connected to the data processing unit or integrated into the data processing unit, the waste objects depicted in the image; and

classifying at least one of the waste objects into one of the predefined waste categories K 1 , K 2 , . . . , K n ;

wherein the waste objects are classified into one of the predefined waste categories K 1 , K 2 , . . . , K n on the basis of a geographical location “P” of the image recording unit or the unseparated waste;

wherein the predefined waste categories comprise at least residual waste, organic waste, paper, plastic, glass, and hazardous waste; wherein the geographical location “P” is provided by a location sensor connected to the data processing unit, and wherein a database, which is connected to the data processing unit or integrated into the data processing unit and in which a relationship between the location “P”, predefined waste objects, and the predefined waste categories K 1 , K 2 , . . . , K n is stored, is used to classify the waste objects into one of the predefined waste categories K 1 , K 2 , . . . , K n .

11 . The device according to claim 10 , wherein the device comprises a mobile phone and wherein the mobile phone comprises a smartphone comprising a user interface and a communication unit.