Direct Profile Modification
The direct manipulation of pixel values or height measurements based on the values of neighboring points is a fundamental method for smoothing paper surface scans. This technique, known as spatial domain filtering, operates directly on the coordinates of the paperboard profile rather than transforming the data into the frequency domain. It allows for the selective removal of localized surface defects or high-frequency noise from measurement profiles.
This process is essential for preparing raw scanner data for roughness calculations and subsequent printability assessments.
Convolution Technique
Applying a moving average or a kernel matrix to the raw data is the primary mechanism of this process. Through spatial domain filtering, each data point is replaced by a weighted average of its neighbors, which smooths out microscopic spikes. This operation helps isolate the structural waviness of the paperboard from the fine surface texture.
It is a necessary step in assessing how the material will behave during high-speed printing and coating operations.
Metrology Limitation
Computational complexity increases when processing high-resolution images of wide sheets. Because each point requires multiple mathematical operations, filtering a full scan in real time can strain processing hardware on fast-moving paper machines. To avoid lag, systems use optimized sub-sampling or parallel processing routines.