Using scale space filtering to make thinning algorithms robust against noise in sketch images

Character Recognition Based on Skeletonization: A Survey.

“…[28] We observed that Skeletonization is a having a huge stride in many image processing applications for few decades. Skeletonization of digitized images is needed due to reasons described in [104]. 5(6), 1503-1519 …”

Skeleton Generation for Digital Images Based on Performance Evaluation Parameters

“…3. To extract important features such as critical end-points, junction-points, and connection among the components can be helpful in many applications [4]. 4.…”

“…(b) Line Following: This method computes centreline by following the contours/edges on either side of the pattern [3,4]. Thus, the skeleton can be formed from connected centrelines.…”

“…Fourier descriptors of the skeleton are obtained and the skeleton can be constructed from a finite set of harmonics [3,4]. (1) 2 < = B(P1) < = 6 (2) A (P1) = 1 (3) At least one of P2 and P4 and P6 is white (4) At least one of P4 and P6 and P8 is white.…”

Sketch-Based Image Retrieval by Size-Adaptive and Noise-Robust Feature Description

“…However, dealing with hand-drawn sketches adds in the challenge of user drawing style variance [11], in addition to the issue of noise that exist usually in sketch images [12].…”