Image Processing Techniques for Braille Writing Recognition

A Review on Software Algorithms for Optical Recognition of Embossed Braille Characters

“…Each time the image was slanted by one pixel in the vertical direction, deviation over the sum of rows was calculated, and a maximum was obtained when the dots were aligned horizontally. Hough transform was used by Antonacopoulos et al [15] for de-skewing the skewed Braille image, while Nestor Falcon et al [17] detected the skew angle of the scanned document by means of horizontal histogram and mass centers calculation and corrected by rotating the original image. A novel binary search algorithm was developed by Abdul Malik S et al [22] to correct various kinds of de-skewing in the scanned images which were tilted.…”

“…For this reason, the partial and local threshold method to finish the segmentation was adopted, namely Braille image could be separated into numerous sub-blocks of appropriate sizes, and then be applied with the Otsu algorithm in each sub-block to segment, and so the sub-block with the size of 60x70 pixels was being utilised. Nestor Falcon et al [17] have used an iterative algorithm which looked for the best threshold according to the areas of Braille dots. This area criterion for thresholds selection offered an accurate way to get the optimum levels to separate black, white and gray level values.…”

“…The proposed system included in it the three major steps namely: half-character detection to determine the whereabouts of the character by detecting the possible dot positions, half-character recognition to determine the half characters, that the dots represent by using a probabilistic neural network and grid determination using transcription algorithm to produce a Braille text file. The approach by Nestor Falcon et al [17] highlights the dot detection using following concepts: dynamic thresholding, pattern detection, Braille grid creation and dot recovery using Braille grid. The research also aimed at applying iterative thresholding algorithms for classifying the grey scale image into black and white and an adaptive algorithm in order to make mesh from the detected dots.…”

Recognition of Double Sided Amharic Braille Documents

“…Accordingly, several OBR systems have been developed over the years for various languages [1,2,3,8]. In general, Braille image recognition systems have four major processes: image acquisition and preprocessing, segmentation, feature extraction and recognition [9]. Braille document images are acquired by digital devices like scanner and this is usually followed by preprocessing with the purpose of enhancing the quality of the Braille image [2,9].…”

“…In general, Braille image recognition systems have four major processes: image acquisition and preprocessing, segmentation, feature extraction and recognition [9]. Braille document images are acquired by digital devices like scanner and this is usually followed by preprocessing with the purpose of enhancing the quality of the Braille image [2,9]. Segmentation involves identification and separation of embossed Braille cells from the background [10].…”