Algorithm for identifying writing stroke and direction

Abstract

Handwriting difficulty is a type of learning disability that may not be detected easily and its diagnosis requires special qualification and experience. Therefore, a new evaluation method is proposed to assist in detecting handwriting problems. This method uses computerized handwriting assessment based on the identification of errors in stroke type, sequences, and direction when forming Latin alphabets. This paper discusses an algorithm to identify type and direction of stroke based on xy-coordinate inputs. The algorithm starts with classification of input into three categories of stroke patterns, which are simple straight line, complex straight line, and curve line. The type and direction of stroke will then be determined by analysis of relationship between consecutive point and also angle difference between points. The algorithm works well in classification and identification involving straight line inputs, while improvements are needed in analyzing curve lines and complex lines involving smooth corner.