A Proposed Scheme for Performance Evaluation of Graphics/Text Separation Algorithms

We propose an objective, comprehensive, and complexity independent metric for performance evaluation of graphics/text separation (text segmentation) algorithms. The metric includes a positive set and a negative set of indices, at both the character and the character string (text) levels, _and it evaluates the detection accuracy of the location, width, height, orientation, skew, string length, and the fragmentation of both characters and strings. Assigning a Segmentation Difficulty (SD) value to the ground truth characters, the performance indices are normalized with respect to the character SD and are therefore independent of the ground truth complexity. The evaluation provides an overall, objective, and comprehensive metric of the text segmentation capability of various algorithms aimed at performing this task.

[1]  Zesheng Tang,et al.  Segmentation and recognition of dimension texts in engineering drawings , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[2]  Osamu Hori,et al.  Quantitative Measurement of the Performance of Raster-to-Vector Conversion Algorithms , 1995, GREC.

[3]  Gao Jingb.  SEGMENTATION AND RECOGNITION OF DIMENSION TEXTS IN ENGINEERING DRAWINGS , 1997 .

[4]  Dov Dori,et al.  A protocol for performance evaluation of line detection algorithms , 1997, Machine Vision and Applications.

[5]  Dov Dori,et al.  Vector-Based Segmentation of Text Connected to Graphics in Engineering Drawings , 1996, SSPR.

[6]  Luigi P. Cordella,et al.  An Alternative Approach to the Performance Evaluation of Thinning Algorithms for Document Processing Applications , 1995, GREC.

[7]  Dov Dori,et al.  Extraction of text boxes from engineering drawings , 1992, Electronic Imaging.

[8]  Rangachar Kasturi,et al.  A Robust Algorithm for Text String Separation from Mixed Text/Graphics Images , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[9]  Dov Dori,et al.  A methodology for the characterization of the performance of thinning algorithms , 1993, Proceedings of 2nd International Conference on Document Analysis and Recognition (ICDAR '93).

[10]  Robert M. Haralick,et al.  Performance characterization in image analysis: thinning, a case in point , 1992, Pattern Recognit. Lett..

[11]  Ching Y. Suen,et al.  Evaluation of thinning algorithms from an OCR viewpoint , 1993, Proceedings of 2nd International Conference on Document Analysis and Recognition (ICDAR '93).

[12]  Karl Tombre,et al.  Graphics Recognition Methods and Applications , 1995, Lecture Notes in Computer Science.

[13]  Robert M. Haralick,et al.  A Benchmark: Performance Evaluation of Dashed-Line Detection Algorithms , 1995, GREC.