From 8c7bf2c380283cf62b7169cbbb25f2fe95465c00 Mon Sep 17 00:00:00 2001 From: Andrey Golovizin Date: Tue, 26 Aug 2014 17:23:07 +0200 Subject: [PATCH] Take optical distance into account when splitting lines. --- pixelocr/page.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/pixelocr/page.py b/pixelocr/page.py index 54d4789..74a2777 100644 --- a/pixelocr/page.py +++ b/pixelocr/page.py @@ -133,7 +133,12 @@ class Page(PageObject): if prev_line is None: prev_line = line else: - if line.top - prev_line.bottom < min_space: + distance = line.top - prev_line.bottom + if ( + # avoid unnecessary calling optical_distance() which may be expensive + distance < min_space + and prev_line.optical_distance(line) < min_space + ): prev_line = Line(self.image[prev_line.top:line.bottom]) else: yield prev_line