Method for extraction of table and figure region from engineering drawings

Patent metadata

Field	Value
Publication number	US-2025191399-A1
Application number	US-202318531747-A
Country	US
Kind code	A1
Filing date	Dec 7, 2023
Priority date	Dec 7, 2023
Publication date	Jun 12, 2025
Grant date	—

Abstract

A system and method of extracting tables and figures from a drawing document is disclosed. The method may include processing coloured image to segmented binary image and extracting a plurality of horizontal lines and a plurality of vertical lines from a foreground of the image. The method may further include detecting a set of candidate table region from the plurality of horizontal lines and the plurality of vertical lines in the image. Further, the method may include calculating textual region density corresponding to each of the set of candidate table regions in the image. The method may further include identifying at least one relevant table region from the set of candidate table regions in the image and a text free region from the at least one additional region in the image. The method may further include identifying at least one figure region from the dilated text free region.

First claim

What is claimed is:

1 . A method of extracting tables and figures from an image, the method comprising:
processing, by an image processing device, the image that corresponds to a segmented binary image for extracting a plurality of horizontal lines and a plurality of vertical lines from a foreground of the image; detecting, by the image processing device, a set of candidate table regions based on intersection of lines from the plurality of horizontal lines and the plurality of vertical lines in the image using a morphological technique; calculating, by the image processing device, textual region density corresponding to each of the set of candidate table regions in the image, based on computation of text area and computation of area of each of the set of candidate table regions; identifying, by the image processing device, at least one relevant table region from the set of candidate table regions in the image, based on the textual region density being above a pre-defined threshold value; identifying, by the image processing device, at least one additional region in the image, based on the textual region density being below or equal to the pre-defined threshold value, wherein the at least one additional region is different from the at least one relevant table region; identifying, by the image processing device, a text free region from the at least one additional region in the image, based on extracting at least one textual region in the at least one additional region, wherein the text free region is dilated using a morphological technique; and identifying, by the image processing device, at least one figure region from the dilated text free region, using a contour-based detection technique.

2 . The method of claim 1 , comprising:
receiving a coloured image from a drawing document; converting the coloured image into a binary image; and segmenting the binary image into the foreground and a background based on an adaptive threshold value.

3 . The method of claim 2 , wherein the binary image is a gray scale image which is black and white in colour.

4 . The method of claim 1 , comprising:
detecting a plurality of table cells from the at least one relevant table region using a contour-based cell detection technique.

5 . The method of claim 1 , wherein the intersection of lines from the plurality of horizontal lines and the plurality of vertical lines in the image is indicative of table cells.

6 . A system for analyzing images and generating a report, the system comprising:
a processor; and a memory communicatively coupled to the processor, wherein the memory stores processor-executable instructions, which, on execution, causes the processor to:
process the image that corresponds to a segmented binary image for extracting a plurality of horizontal lines and a plurality of vertical lines from a foreground of the image;
detect a set of candidate table regions based on intersection of lines from the plurality of horizontal lines and the plurality of vertical lines in the image using a morphological technique;
calculate textual region density corresponding to each of the set of candidate table regions in the image, based on computation of text area and computation of area of each of the set of candidate table regions;
identify at least one relevant table region from the set of candidate table regions in the image, based on the textual region density being above a pre-defined threshold value;
identify at least one additional region in the image, based on the textual region density being below or equal to the pre-defined threshold value, wherein the at least one additional region is different from the at least one relevant table region;
identify a text free region from the at least one additional region in the image, based on extracting at least one textual region in the at least one additional region, wherein the text free region is dilated using a morphological technique; and
identify at least one figure region from the dilated text free region, using a contour-based detection technique.

7 . The system of claim 6 , wherein the processor-executable instructions further cause the processor to:
receiving a coloured image from a drawing document; converting the coloured image into a binary image; and
segmenting the binary image into the foreground and a background based on an adaptive threshold value.

8 . The system of claim 7 , wherein the binary image is a gray scale image which is black and white in colour.

9 . The system of claim 6 , wherein the processor-executable instructions further cause the processor to detect a plurality of table cells from the at least one relevant table region using a contour-based cell detection technique.

10 . The system of claim 6 , wherein the intersection of lines from the plurality of horizontal lines and the plurality of vertical lines in the image is indicative of table cells.

11 . A non-transitory computer-readable medium storing computer-executable instructions for analysing images and generating a report, the computer-executable instructions configured for:
processing the image that corresponds to a segmented binary image for extracting a plurality of horizontal lines and a plurality of vertical lines from a foreground of the image; detecting a set of candidate table regions based on intersection of lines from the plurality of horizontal lines and the plurality of vertical lines in the image using a morphological technique; calculating textual region density corresponding to each of the set of candidate table regions in the image, based on computation of text area and computation of area of each of the set of candidate table regions; identifying at least one relevant table region from the set of candidate table regions in the image, based on the textual region density being above a pre-defined threshold value; identifying at least one additional region in the image, based on the textual region density being below or equal to the pre-defined threshold value, wherein the at least one additional region is different from the at least one relevant table region; identifying a text free region from the at least one additional region in the image, based on extracting at least one textual region in the at least one additional region, wherein the text free region is dilated using a morphological technique; and identifying at least one figure region from the dilated text free region, using a contour-based detection technique.

12 . The non-transitory computer-readable medium of claim 11 , wherein the computer-executable instructions are configured for:
receiving a coloured image from a drawing document; converting the coloured image into a binary image; and segmenting the binary image into the foreground and a background based on an adaptive threshold value.

13 . The non-transitory computer-readable medium of claim 12 , wherein the binary image is a grayscale image which is black and white in colour.

14 . The non-transitory computer-readable medium of claim 11 , wherein the computer-executable instructions are configured for:
detecting a plurality of table cells from the at least one relevant table region using a contour-based cell detection technique.

15 . The non-transitory computer-readable medium of claim 11 , wherein the intersection of lines from the plurality of horizontal lines and the plurality of vertical lines in the image is indicative of table cells.

Assignees

• L&T Technology Services Ltd

Inventors

• Mallick Triptesh
• Das Tarun Kumar
• Balaraman Mridul
• Singh Madhusudan

Classifications

• G06V30/18152

  Extracting features based on a plurality of salient regional features, e.g. "bag of words" · CPC title

• G06T7/11

  Region-based segmentation · CPC title

• G06V30/412

  Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables · CPC title

• G06T7/155

  involving morphological operators · CPC title

• G06V30/414 Primary

  Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text · CPC title