Segmentation for touching Thai typewrittens
Sarin Watcharabutsarakham · 2004
Thai characters vary in size and position in sentences and Thai word is composed of consonants, vowels and tones. Touching characters can occur both horizontal and vertical directions. This paper proposes a segmentation technique for touching Thai typewrittens. The proposal uses structural characteristics to detect suitable segmentation points in both horizontal and vertical directions. The segmentation process consists of 4 steps: line classification, zone identification, touching type and segmentation type. The proposed segmentation technique tested both electronic typewriters and manual portable typewriters. The segmentation accuracy of 94% has been obtained for 100 sentences of typewritten thesis documents.