Cursive Handwritten Segmentation and Recognition for Instructional Videos

摘要: In this paper, we address the issues pertaining to segmentation and recognition of cursive handwritten text from chalkboard lecture videos. Recognizing is a challenging problem in instructor-led video. The task gets even tougher with varying handwriting styles blackboard type. Unlike on whiteboard electronic boards, represents serious challenges such as, lack uniform edge density, weak chalk contrast against leftover dust noise as result erasing -- many others. Moreover, color boards illumination changes within video makes it impossible use trivial thresholding techniques, for extraction content. Many universities throughout world still heavily rely mode instruction. Therefore, recognizing these content will not only aid indexing retrieval applications but also help understand high level semantics, useful Multi-media Learning Objects (MLO). order encounter those adversaries, here propose system We first create foreground model segment background blackboard. then characters using one-dimensional vertical histogram. Later, extract gradient based features classify an SVM classifier. obtained encouraging accuracy 86.28% 5-fold cross validation.