Automatic Image Analysis for Processing Marks in Femtosecond Laser Micromachining Using Concave and Convex (unevenness) Coefficient

In this paper, we demonstrate automatic image analysis for processing marks in femtosecond laser micromachining using concave and convex (unevenness) coefficient. To realize laser processing without thermal deformation and cracks, it's important to search optimum processing conditions such as scanning speed and pulse energy. Therefore, we analyzed a set of microscope image during laser microprocessing of glass, and evaluated the morphology such as depth and straightness of processing marks. For the laser microprocessing, an ultrashort laser system (Fianium, FP1060S-PP-D) with a wavelength of 1.06 mu m, a pulse duration of 250 fs, and a repetition rate of 1 MHz was used. The sample to be processed was white glass substrates. The sample, which was mounted on the threedimensional stage, was scanned 10 mm in the x-axis direction with a scan speed of 0.1, 0.5, 1, and 2 mm/s. Also, for the laser microprocessing, the pulse energy was changed from 0.9 to 1.3 mu J. After laser microprocessing, processed mark was analyzed with concave and convex coefficient and Otsu's method. By making use of the concave and convex coefficient, luminance unevenness was reduced. We will introduce relationship between "scanning speed and/or pulse energy" and "gray scale and/or straightness of processing marks".