def splitToSpaces(line): line = xycuts.cut_edges(line) central_ind = _central_ind(line) central_spaces = _find_central_spaces(line, central_ind) straight_spaces, other_spaces = _filter_straight_spaces(line, central_spaces) word_spaces, symbol_spaces = _filter_word_spaces(line, straight_spaces) return word_spaces, symbol_spaces, other_spaces
def doSplit(line): """split given zone (seems to be a line of text) and returns list of words each one interpreted""" line = xycuts.cut_edges(line) central_ind = _central_ind(line) central_spaces = _find_central_spaces(line, central_ind) straight_spaces, other_spaces = _filter_straight_spaces(line, central_spaces) word_spaces, symbol_spaces = _filter_word_spaces(line, straight_spaces) wry_slices = _filter_wry_spaces(line, central_ind, other_spaces) return _makeWords(line, word_spaces, symbol_spaces, wry_slices)
def makeWords(self, word_spaces): start = self.line.horizontal[0] words = [] for space in word_spaces: end = space[0] words.append(WordSpace(self.line, (start, end))) start = space[1] end = self.line.horizontal[1] words.append(WordSpace(self.line, (start, end))) return words im = readGrayIm("../images/line_0.bmp") im[im < 0.3] = 0 zone = sspace.Zone(im) zone = xycuts.cut_edges(zone) central_ind = sspace._central_ind(zone) line = sspace.Line(zone.im, central_ind, horizontal=zone.horizontal, vertical=zone.vertical) word_spaces = Splitter(line).find_word_spaces() word_id = 0 output = "../output/" rmtree(output) mkdir(output) for word_space in word_spaces: word = word_space.toWord() writeGrayIm(output + "word_%s.bmp" % word_id, word.extract()) symbol_id = 0 mkdir(output + "word_%s" % word_id) for symbol in word.symbols: writeGrayIm(output + "word_%s/symbol_%s.bmp" % (word_id, symbol_id), symbol.extract()) symbol_id += 1
def makeWords(self, word_spaces): start = self.line.horizontal[0] words = [] for space in word_spaces: end = space[0] words.append(WordSpace(self.line, (start, end))) start = space[1] end = self.line.horizontal[1] words.append(WordSpace(self.line, (start, end))) return words im = readGrayIm("../images/line_0.bmp") im[im < 0.3] = 0 zone = sspace.Zone(im) zone = xycuts.cut_edges(zone) central_ind = sspace._central_ind(zone) line = sspace.Line(zone.im, central_ind, horizontal=zone.horizontal, vertical=zone.vertical) word_spaces = Splitter(line).find_word_spaces() word_id = 0 output = "../output/" rmtree(output) mkdir(output) for word_space in word_spaces: word = word_space.toWord() writeGrayIm(output + "word_%s.bmp" % word_id, word.extract()) symbol_id = 0 mkdir(output + "word_%s" % word_id)