During the assembly of a control cabinet, a worker obstructs many individual configured wires. To distinguish these wires, a printer plots an identifying text on each end of the wires. However, due to the shape of the wires and the printing process, the quality of these markings is often too low, and it is hard or impossible to read the marking. Common reasons are a low contrast or a blurred text. By now, there is no quality check of the marking after a crimping machine produced the wire. This paper investigates methods for wire mark reading that is required to estimate the quality of the marking. By using optical character recognition, the likeliness that a worker can read the marking must be computed. In the final solution, the quality check of the marking will be implemented within an automated quality check that is located after the printing process. With this, the crimping machine can then discard wires of low quality and reproduce them instantly.