Morphological preprocessing method to thresholding degraded word images Export

@article{citeulike:4281154, abstract = {This paper presents a novel preprocessing method based on mathematical morphology techniques to improve the subsequent thresholding quality of raw degraded word images. The raw degraded word images contain undesirable shapes called critical shadows on the background that cause noise in binary images. This noise constitutes obstacles to posterior segmentation of characters. Direct application of a thresholding method produces inadequate binary versions of these degraded word images. Our preprocessing method called Shadow Location and Lightening (SL*L) adaptively, accurately and without manual fine-tuning of parameters locates these critical shadows on grayscale degraded images using morphological operations, and lightens them before applying eventual thresholding process. In this way, enhanced binary images without unpredictable and inappropriate noise can be provided to subsequent segmentation of characters. Then, adequate binary characters can be segmented and extracted as input data to optical character recognition (OCR) applications saving computational effort and increasing recognition rate. The proposed method is experimentally tested with a set of several raw degraded images extracted from real photos acquired by unsophisticated imaging systems. A qualitative analysis of experimental results led to conclusions that the thresholding result quality was significantly improved with the proposed preprocessing method. Also, a quantitative evaluation using a testing data of 1194 degraded word images showed the essentiality and effectiveness of the proposed preprocessing method to increase segmentation and recognition rates of their characters. Furthermore, an advantage of the proposed method is that Otsu's method as a simple and easily implementable global thresholding technique can be sufficient to reducing computational load.}, author = {Nomura, Shigueo and Yamanaka, Keiji and Shiose, Takayuki and Kawakami, Hiroshi and Katai, Osamu}, citeulike-article-id = {4281154}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/j.patrec.2009.03.008}, citeulike-linkout-1 = {http://linkinghub.elsevier.com/retrieve/pii/S016786550900049X}, comment = {* sample images look like bibs * found from Keith Price's bibliography http://www.visionbib.com/bibliography/char981.html * uses thickening+pruning to detect shadows (thickened images look like skeletonized images) * some heuristic to handle upper and lower shadows separately * written MS thesis on binarization in 2002 (in Portuguese)}, day = {01}, doi = {10.1016/j.patrec.2009.03.008}, issn = {01678655}, journal = {Pattern Recognition Letters}, keywords = {ocr, preprocessing}, month = {June}, number = {8}, pages = {729--744}, posted-at = {2009-07-12 22:47:50}, priority = {2}, title = {Morphological preprocessing method to thresholding degraded word images}, url = {http://dx.doi.org/10.1016/j.patrec.2009.03.008}, volume = {30}, year = {2009} }

TY - JOUR ID - citeulike:4281154 L3 - citeulike-article-id:4281154 N2 - This paper presents a novel preprocessing method based on mathematical morphology techniques to improve the subsequent thresholding quality of raw degraded word images. The raw degraded word images contain undesirable shapes called critical shadows on the background that cause noise in binary images. This noise constitutes obstacles to posterior segmentation of characters. Direct application of a thresholding method produces inadequate binary versions of these degraded word images. Our preprocessing method called Shadow Location and Lightening (SL*L) adaptively, accurately and without manual fine-tuning of parameters locates these critical shadows on grayscale degraded images using morphological operations, and lightens them before applying eventual thresholding process. In this way, enhanced binary images without unpredictable and inappropriate noise can be provided to subsequent segmentation of characters. Then, adequate binary characters can be segmented and extracted as input data to optical character recognition (OCR) applications saving computational effort and increasing recognition rate. The proposed method is experimentally tested with a set of several raw degraded images extracted from real photos acquired by unsophisticated imaging systems. A qualitative analysis of experimental results led to conclusions that the thresholding result quality was significantly improved with the proposed preprocessing method. Also, a quantitative evaluation using a testing data of 1194 degraded word images showed the essentiality and effectiveness of the proposed preprocessing method to increase segmentation and recognition rates of their characters. Furthermore, an advantage of the proposed method is that Otsu’s method as a simple and easily implementable global thresholding technique can be sufficient to reducing computational load. IS - 8 JF - Pattern Recognition Letters SN - 01678655 EP - 744 TI - Morphological preprocessing method to thresholding degraded word images VL - 30 SP - 729 KW - ocr KW - preprocessing N1 - * sample images look like bibs * found from Keith Price's bibliography http://www.visionbib.com/bibliography/char981.html * uses thickening+pruning to detect shadows (thickened images look like skeletonized images) * some heuristic to handle upper and lower shadows separately * written MS thesis on binarization in 2002 (in Portuguese) AU - Nomura, Shigueo AU - Yamanaka, Keiji AU - Shiose, Takayuki AU - Kawakami, Hiroshi AU - Katai, Osamu PY - 2009/06/01/ UR - http://dx.doi.org/10.1016/j.patrec.2009.03.008 ER -