Automated two-dimensional patterned shape cutting of elastic web materials

The objective of this thesis is to automate the manual 2D-patterned cutting of lace with high a accuracy and quality of the cutting edges. Four main problems for automating the cutting method were addressed; feature recognition technique for cutting path extraction, material handling system, laser beam manipulation and trajectory planning generation. An integrated mechatronic approach for designing the automated laser cutting system was outlined. The features of the 2D-patterned lace fabrics are not exactly identical and easily distorted nonlinearly due to handling operations. An active template matching technique was developed and implemented to recognise the distorted features of the 2D-patterned shape cuttings. To accommodate the geometrical pattern variations, the cutting paths of the 2D-patterned shapes were extracted using either the chain-coding technique or the binary morphological techniques. [Continues.]