The thesis discusses the possibility for the use of a computer algorithm to perform laser cut analysis from the picture of a structured test sample, as an equivalent method to the classic manual applicative cut analysis. The applicative laser cut analysis is a popular method because of its simplicity and low cost, but it also has some drawbacks, such as experience requirements of the operator and subjective result interpretation. The mentioned drawbacks of the manual applicative method can be avoided with the use of the algorithm analysis, which was designed for this task. The algorithm analysis is a fast and objective process of interpreting results and it does not require a high level of experience from the operator.
The first chapter describes laser structuring, laser light properties and existing methods of determining laser focus position. The applicative method of laser cut analysis and the influence of laser source and structuring parameters on the structured sample are described in the second chapter. The third chapter describes the algorithm analysis, which is performed on a picture of a specifically designed test sample. The test sample has a specific structure, from which the searched parameters values can be found. The algorithm can perform a rough beam focus position estimation, a fine beam focus estimation with astigmatism estimation and maximum structuring speed estimation.
The algorithm was tested on multiple laser sources and the test results were compared with the results of the manual applicative method in the fourth chapter of the thesis. Additional testing was performed to assess the analysis repeatability, test sample lighting influence, material position change and structuring speed influence on the analysis results. Test results proved that the algorithm analysis is perfectly comparable to the manual applicative method results.