Method for determining a sharpness condition of a saw chain of a chain saw

A method for determining a sharpness condition of a saw chain of a chain saw, wherein the saw chain includes at least one cutting link having an upper cutting blade, wherein the upper cutting blade has a width of less than 500 micrometers, wherein an upper cutting blade said to be sharp has a width of the method comprising:

• providing a reference saw chain defined by a sharp saw chain having an upper cutting blade having a width of less than 50 μm;
• recording an upper cutting blade image of the upper cutting blade of the cutting link via an imaging device;
• performing an evaluation of the upper cutting blade image using an evaluation unit, wherein the evaluation unit includes a first artificial neural network, a sharp condition of the saw chain is defined in the evaluation unit, a dull condition of the saw chain is defined in the evaluation unit, and the saw chain is assigned to the sharp condition or the dull condition on a basis of the upper cutting blade image using the first artificial neural network, wherein the first artificial neural network uses a data set of digital images of upper cutting blades of sharp saw chains and of upper cutting blades of dull saw chains in order to decide whether the saw chain is to be assigned to the dull condition or to the sharp condition;
• wherein a sharp saw chain of the saw chains of the data set supplies at least 80% of the cutting performance of a reference saw chain in identical working conditions;
• wherein a dull saw chain supplies at most 65% of the cutting performance of the reference saw chain in identical working conditions;
• the sharp saw chains are assigned to the sharp condition;
• and, the dull saw chains are assigned to the dull condition, wherein the evaluation unit is configured to output a directive on a basis of the sharpness condition of the saw chain, wherein the directive includes an instruction to resharpen the saw chain;
• and,
• wherein the evaluation unit includes at least one second artificial neural network;
• the second artificial neural network is a convolutional neural network and is trained on recognition of upper cutting blades in upper cutting blade images;
• wherein, with the aid of the second artificial neural network before the assignment of the upper cutting blade image to the sharp condition or the dull condition via the first artificial neural network, a region of the upper cutting blade image in which the upper cutting blade is located is selected;
• and, the upper cutting blade image is cropped to the region.